refactor: prefer checks against nullrev over nullid

A common pattern is using a changeset context and obtaining the node to
compare against nullid. Change this to obtain the nullrev instead. In
the future, the nullid becomes a property of the repository and is no
longer a global constant, so using nullrev is much easier to reason
about. Python function call overhead makes the difference moot, but
future changes will result in more dictionary lookups otherwise, so
prefer the simpler pattern.

Differential Revision: https://phab.mercurial-scm.org/D10290

hgext/extdiff.py

@@ -1,803 +1,803 b''
 # extdiff.py - external diff program support for mercurial
 #
 # Copyright 2006 Vadim Gelfer <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 '''command to allow external programs to compare revisions
 
 The extdiff Mercurial extension allows you to use external programs
 to compare revisions, or revision with working directory. The external
 diff programs are called with a configurable set of options and two
 non-option arguments: paths to directories containing snapshots of
 files to compare.
 
 If there is more than one file being compared and the "child" revision
 is the working directory, any modifications made in the external diff
 program will be copied back to the working directory from the temporary
 directory.
 
 The extdiff extension also allows you to configure new diff commands, so
 you do not need to type :hg:`extdiff -p kdiff3` always. ::
 
   [extdiff]
   # add new command that runs GNU diff(1) in 'context diff' mode
   cdiff = gdiff -Nprc5
   ## or the old way:
   #cmd.cdiff = gdiff
   #opts.cdiff = -Nprc5
 
   # add new command called meld, runs meld (no need to name twice).  If
   # the meld executable is not available, the meld tool in [merge-tools]
   # will be used, if available
   meld =
 
   # add new command called vimdiff, runs gvimdiff with DirDiff plugin
   # (see http://www.vim.org/scripts/script.php?script_id=102) Non
   # English user, be sure to put "let g:DirDiffDynamicDiffText = 1" in
   # your .vimrc
   vimdiff = gvim -f "+next" \\
             "+execute 'DirDiff' fnameescape(argv(0)) fnameescape(argv(1))"
 
 Tool arguments can include variables that are expanded at runtime::
 
   $parent1, $plabel1 - filename, descriptive label of first parent
   $child,   $clabel  - filename, descriptive label of child revision
   $parent2, $plabel2 - filename, descriptive label of second parent
   $root              - repository root
   $parent is an alias for $parent1.
 
 The extdiff extension will look in your [diff-tools] and [merge-tools]
 sections for diff tool arguments, when none are specified in [extdiff].
 
 ::
 
   [extdiff]
   kdiff3 =
 
   [diff-tools]
   kdiff3.diffargs=--L1 '$plabel1' --L2 '$clabel' $parent $child
 
 If a program has a graphical interface, it might be interesting to tell
 Mercurial about it. It will prevent the program from being mistakenly
 used in a terminal-only environment (such as an SSH terminal session),
 and will make :hg:`extdiff --per-file` open multiple file diffs at once
 instead of one by one (if you still want to open file diffs one by one,
 you can use the --confirm option).
 
 Declaring that a tool has a graphical interface can be done with the
 ``gui`` flag next to where ``diffargs`` are specified:
 
 ::
 
   [diff-tools]
   kdiff3.diffargs=--L1 '$plabel1' --L2 '$clabel' $parent $child
   kdiff3.gui = true
 
 You can use -I/-X and list of file or directory names like normal
 :hg:`diff` command. The extdiff extension makes snapshots of only
 needed files, so running the external diff program will actually be
 pretty fast (at least faster than having to compare the entire tree).
 '''
 
 from __future__ import absolute_import
 
 import os
 import re
 import shutil
 import stat
 import subprocess
 
 from mercurial.i18n import _
 from mercurial.node import (
-    nullid,
+    nullrev,
     short,
 )
 from mercurial import (
     archival,
     cmdutil,
     encoding,
     error,
     filemerge,
     formatter,
     pycompat,
     registrar,
     scmutil,
     util,
 )
 from mercurial.utils import (
     procutil,
     stringutil,
 )
 
 cmdtable = {}
 command = registrar.command(cmdtable)
 
 configtable = {}
 configitem = registrar.configitem(configtable)
 
 configitem(
     b'extdiff',
     br'opts\..*',
     default=b'',
     generic=True,
 )
 
 configitem(
     b'extdiff',
     br'gui\..*',
     generic=True,
 )
 
 configitem(
     b'diff-tools',
     br'.*\.diffargs$',
     default=None,
     generic=True,
 )
 
 configitem(
     b'diff-tools',
     br'.*\.gui$',
     generic=True,
 )
 
 # Note for extension authors: ONLY specify testedwith = 'ships-with-hg-core' for
 # extensions which SHIP WITH MERCURIAL. Non-mainline extensions should
 # be specifying the version(s) of Mercurial they are tested with, or
 # leave the attribute unspecified.
 testedwith = b'ships-with-hg-core'
 
 
 def snapshot(ui, repo, files, node, tmproot, listsubrepos):
     """snapshot files as of some revision
     if not using snapshot, -I/-X does not work and recursive diff
     in tools like kdiff3 and meld displays too many files."""
     dirname = os.path.basename(repo.root)
     if dirname == b"":
         dirname = b"root"
     if node is not None:
         dirname = b'%s.%s' % (dirname, short(node))
     base = os.path.join(tmproot, dirname)
     os.mkdir(base)
     fnsandstat = []
 
     if node is not None:
         ui.note(
             _(b'making snapshot of %d files from rev %s\n')
             % (len(files), short(node))
         )
     else:
         ui.note(
             _(b'making snapshot of %d files from working directory\n')
             % (len(files))
         )
 
     if files:
         repo.ui.setconfig(b"ui", b"archivemeta", False)
 
         archival.archive(
             repo,
             base,
             node,
             b'files',
             match=scmutil.matchfiles(repo, files),
             subrepos=listsubrepos,
         )
 
         for fn in sorted(files):
             wfn = util.pconvert(fn)
             ui.note(b'  %s\n' % wfn)
 
             if node is None:
                 dest = os.path.join(base, wfn)
 
                 fnsandstat.append((dest, repo.wjoin(fn), os.lstat(dest)))
     return dirname, fnsandstat
 
 
 def formatcmdline(
     cmdline,
     repo_root,
     do3way,
     parent1,
     plabel1,
     parent2,
     plabel2,
     child,
     clabel,
 ):
     # Function to quote file/dir names in the argument string.
     # When not operating in 3-way mode, an empty string is
     # returned for parent2
     replace = {
         b'parent': parent1,
         b'parent1': parent1,
         b'parent2': parent2,
         b'plabel1': plabel1,
         b'plabel2': plabel2,
         b'child': child,
         b'clabel': clabel,
         b'root': repo_root,
     }
 
     def quote(match):
         pre = match.group(2)
         key = match.group(3)
         if not do3way and key == b'parent2':
             return pre
         return pre + procutil.shellquote(replace[key])
 
     # Match parent2 first, so 'parent1?' will match both parent1 and parent
     regex = (
         br'''(['"]?)([^\s'"$]*)'''
         br'\$(parent2|parent1?|child|plabel1|plabel2|clabel|root)\1'
     )
     if not do3way and not re.search(regex, cmdline):
         cmdline += b' $parent1 $child'
     return re.sub(regex, quote, cmdline)
 
 
 def _systembackground(cmd, environ=None, cwd=None):
     """like 'procutil.system', but returns the Popen object directly
     so we don't have to wait on it.
     """
     env = procutil.shellenviron(environ)
     proc = subprocess.Popen(
         procutil.tonativestr(cmd),
         shell=True,
         close_fds=procutil.closefds,
         env=procutil.tonativeenv(env),
         cwd=pycompat.rapply(procutil.tonativestr, cwd),
     )
     return proc
 
 
 def _runperfilediff(
     cmdline,
     repo_root,
     ui,
     guitool,
     do3way,
     confirm,
     commonfiles,
     tmproot,
     dir1a,
     dir1b,
     dir2,
     rev1a,
     rev1b,
     rev2,
 ):
     # Note that we need to sort the list of files because it was
     # built in an "unstable" way and it's annoying to get files in a
     # random order, especially when "confirm" mode is enabled.
     waitprocs = []
     totalfiles = len(commonfiles)
     for idx, commonfile in enumerate(sorted(commonfiles)):
         path1a = os.path.join(dir1a, commonfile)
         label1a = commonfile + rev1a
         if not os.path.isfile(path1a):
             path1a = pycompat.osdevnull
 
         path1b = b''
         label1b = b''
         if do3way:
             path1b = os.path.join(dir1b, commonfile)
             label1b = commonfile + rev1b
             if not os.path.isfile(path1b):
                 path1b = pycompat.osdevnull
 
         path2 = os.path.join(dir2, commonfile)
         label2 = commonfile + rev2
 
         if confirm:
             # Prompt before showing this diff
             difffiles = _(b'diff %s (%d of %d)') % (
                 commonfile,
                 idx + 1,
                 totalfiles,
             )
             responses = _(
                 b'[Yns?]'
                 b'$$ &Yes, show diff'
                 b'$$ &No, skip this diff'
                 b'$$ &Skip remaining diffs'
                 b'$$ &? (display help)'
             )
             r = ui.promptchoice(b'%s %s' % (difffiles, responses))
             if r == 3:  # ?
                 while r == 3:
                     for c, t in ui.extractchoices(responses)[1]:
                         ui.write(b'%s - %s\n' % (c, encoding.lower(t)))
                     r = ui.promptchoice(b'%s %s' % (difffiles, responses))
             if r == 0:  # yes
                 pass
             elif r == 1:  # no
                 continue
             elif r == 2:  # skip
                 break
 
         curcmdline = formatcmdline(
             cmdline,
             repo_root,
             do3way=do3way,
             parent1=path1a,
             plabel1=label1a,
             parent2=path1b,
             plabel2=label1b,
             child=path2,
             clabel=label2,
         )
 
         if confirm or not guitool:
             # Run the comparison program and wait for it to exit
             # before we show the next file.
             # This is because either we need to wait for confirmation
             # from the user between each invocation, or because, as far
             # as we know, the tool doesn't have a GUI, in which case
             # we can't run multiple CLI programs at the same time.
             ui.debug(
                 b'running %r in %s\n' % (pycompat.bytestr(curcmdline), tmproot)
             )
             ui.system(curcmdline, cwd=tmproot, blockedtag=b'extdiff')
         else:
             # Run the comparison program but don't wait, as we're
             # going to rapid-fire each file diff and then wait on
             # the whole group.
             ui.debug(
                 b'running %r in %s (backgrounded)\n'
                 % (pycompat.bytestr(curcmdline), tmproot)
             )
             proc = _systembackground(curcmdline, cwd=tmproot)
             waitprocs.append(proc)
 
     if waitprocs:
         with ui.timeblockedsection(b'extdiff'):
             for proc in waitprocs:
                 proc.wait()
 
 
 def diffpatch(ui, repo, node1, node2, tmproot, matcher, cmdline):
     template = b'hg-%h.patch'
     # write patches to temporary files
     with formatter.nullformatter(ui, b'extdiff', {}) as fm:
         cmdutil.export(
             repo,
             [repo[node1].rev(), repo[node2].rev()],
             fm,
             fntemplate=repo.vfs.reljoin(tmproot, template),
             match=matcher,
         )
     label1 = cmdutil.makefilename(repo[node1], template)
     label2 = cmdutil.makefilename(repo[node2], template)
     file1 = repo.vfs.reljoin(tmproot, label1)
     file2 = repo.vfs.reljoin(tmproot, label2)
     cmdline = formatcmdline(
         cmdline,
         repo.root,
         # no 3way while comparing patches
         do3way=False,
         parent1=file1,
         plabel1=label1,
         # while comparing patches, there is no second parent
         parent2=None,
         plabel2=None,
         child=file2,
         clabel=label2,
     )
     ui.debug(b'running %r in %s\n' % (pycompat.bytestr(cmdline), tmproot))
     ui.system(cmdline, cwd=tmproot, blockedtag=b'extdiff')
     return 1
 
 
 def diffrevs(
     ui,
     repo,
     ctx1a,
     ctx1b,
     ctx2,
     matcher,
     tmproot,
     cmdline,
     do3way,
     guitool,
     opts,
 ):
 
     subrepos = opts.get(b'subrepos')
 
     # calculate list of files changed between both revs
     st = ctx1a.status(ctx2, matcher, listsubrepos=subrepos)
     mod_a, add_a, rem_a = set(st.modified), set(st.added), set(st.removed)
     if do3way:
         stb = ctx1b.status(ctx2, matcher, listsubrepos=subrepos)
         mod_b, add_b, rem_b = (
             set(stb.modified),
             set(stb.added),
             set(stb.removed),
         )
     else:
         mod_b, add_b, rem_b = set(), set(), set()
     modadd = mod_a | add_a | mod_b | add_b
     common = modadd | rem_a | rem_b
     if not common:
         return 0
 
     # Always make a copy of ctx1a (and ctx1b, if applicable)
     # dir1a should contain files which are:
     #   * modified or removed from ctx1a to ctx2
     #   * modified or added from ctx1b to ctx2
     #     (except file added from ctx1a to ctx2 as they were not present in
     #     ctx1a)
     dir1a_files = mod_a | rem_a | ((mod_b | add_b) - add_a)
     dir1a = snapshot(ui, repo, dir1a_files, ctx1a.node(), tmproot, subrepos)[0]
     rev1a = b'' if ctx1a.rev() is None else b'@%d' % ctx1a.rev()
     if do3way:
         # file calculation criteria same as dir1a
         dir1b_files = mod_b | rem_b | ((mod_a | add_a) - add_b)
         dir1b = snapshot(
             ui, repo, dir1b_files, ctx1b.node(), tmproot, subrepos
         )[0]
         rev1b = b'@%d' % ctx1b.rev()
     else:
         dir1b = None
         rev1b = b''
 
     fnsandstat = []
 
     # If ctx2 is not the wc or there is >1 change, copy it
     dir2root = b''
     rev2 = b''
     if ctx2.node() is not None:
         dir2 = snapshot(ui, repo, modadd, ctx2.node(), tmproot, subrepos)[0]
         rev2 = b'@%d' % ctx2.rev()
     elif len(common) > 1:
         # we only actually need to get the files to copy back to
         # the working dir in this case (because the other cases
         # are: diffing 2 revisions or single file -- in which case
         # the file is already directly passed to the diff tool).
         dir2, fnsandstat = snapshot(ui, repo, modadd, None, tmproot, subrepos)
     else:
         # This lets the diff tool open the changed file directly
         dir2 = b''
         dir2root = repo.root
 
     label1a = rev1a
     label1b = rev1b
     label2 = rev2
 
     if not opts.get(b'per_file'):
         # If only one change, diff the files instead of the directories
         # Handle bogus modifies correctly by checking if the files exist
         if len(common) == 1:
             common_file = util.localpath(common.pop())
             dir1a = os.path.join(tmproot, dir1a, common_file)
             label1a = common_file + rev1a
             if not os.path.isfile(dir1a):
                 dir1a = pycompat.osdevnull
             if do3way:
                 dir1b = os.path.join(tmproot, dir1b, common_file)
                 label1b = common_file + rev1b
                 if not os.path.isfile(dir1b):
                     dir1b = pycompat.osdevnull
             dir2 = os.path.join(dir2root, dir2, common_file)
             label2 = common_file + rev2
 
         # Run the external tool on the 2 temp directories or the patches
         cmdline = formatcmdline(
             cmdline,
             repo.root,
             do3way=do3way,
             parent1=dir1a,
             plabel1=label1a,
             parent2=dir1b,
             plabel2=label1b,
             child=dir2,
             clabel=label2,
         )
         ui.debug(b'running %r in %s\n' % (pycompat.bytestr(cmdline), tmproot))
         ui.system(cmdline, cwd=tmproot, blockedtag=b'extdiff')
     else:
         # Run the external tool once for each pair of files
         _runperfilediff(
             cmdline,
             repo.root,
             ui,
             guitool=guitool,
             do3way=do3way,
             confirm=opts.get(b'confirm'),
             commonfiles=common,
             tmproot=tmproot,
             dir1a=os.path.join(tmproot, dir1a),
             dir1b=os.path.join(tmproot, dir1b) if do3way else None,
             dir2=os.path.join(dir2root, dir2),
             rev1a=rev1a,
             rev1b=rev1b,
             rev2=rev2,
         )
 
     for copy_fn, working_fn, st in fnsandstat:
         cpstat = os.lstat(copy_fn)
         # Some tools copy the file and attributes, so mtime may not detect
         # all changes.  A size check will detect more cases, but not all.
         # The only certain way to detect every case is to diff all files,
         # which could be expensive.
         # copyfile() carries over the permission, so the mode check could
         # be in an 'elif' branch, but for the case where the file has
         # changed without affecting mtime or size.
         if (
             cpstat[stat.ST_MTIME] != st[stat.ST_MTIME]
             or cpstat.st_size != st.st_size
             or (cpstat.st_mode & 0o100) != (st.st_mode & 0o100)
         ):
             ui.debug(
                 b'file changed while diffing. '
                 b'Overwriting: %s (src: %s)\n' % (working_fn, copy_fn)
             )
             util.copyfile(copy_fn, working_fn)
 
     return 1
 
 
 def dodiff(ui, repo, cmdline, pats, opts, guitool=False):
     """Do the actual diff:
 
     - copy to a temp structure if diffing 2 internal revisions
     - copy to a temp structure if diffing working revision with
       another one and more than 1 file is changed
     - just invoke the diff for a single file in the working dir
     """
 
     cmdutil.check_at_most_one_arg(opts, b'rev', b'change')
     revs = opts.get(b'rev')
     from_rev = opts.get(b'from')
     to_rev = opts.get(b'to')
     change = opts.get(b'change')
     do3way = b'$parent2' in cmdline
 
     if change:
         ctx2 = scmutil.revsingle(repo, change, None)
         ctx1a, ctx1b = ctx2.p1(), ctx2.p2()
     elif from_rev or to_rev:
         repo = scmutil.unhidehashlikerevs(
             repo, [from_rev] + [to_rev], b'nowarn'
         )
         ctx1a = scmutil.revsingle(repo, from_rev, None)
-        ctx1b = repo[nullid]
+        ctx1b = repo[nullrev]
         ctx2 = scmutil.revsingle(repo, to_rev, None)
     else:
         ctx1a, ctx2 = scmutil.revpair(repo, revs)
         if not revs:
             ctx1b = repo[None].p2()
         else:
-            ctx1b = repo[nullid]
+            ctx1b = repo[nullrev]
 
     # Disable 3-way merge if there is only one parent
     if do3way:
-        if ctx1b.node() == nullid:
+        if ctx1b.rev() == nullrev:
             do3way = False
 
     matcher = scmutil.match(ctx2, pats, opts)
 
     if opts.get(b'patch'):
         if opts.get(b'subrepos'):
             raise error.Abort(_(b'--patch cannot be used with --subrepos'))
         if opts.get(b'per_file'):
             raise error.Abort(_(b'--patch cannot be used with --per-file'))
         if ctx2.node() is None:
             raise error.Abort(_(b'--patch requires two revisions'))
 
     tmproot = pycompat.mkdtemp(prefix=b'extdiff.')
     try:
         if opts.get(b'patch'):
             return diffpatch(
                 ui, repo, ctx1a.node(), ctx2.node(), tmproot, matcher, cmdline
             )
 
         return diffrevs(
             ui,
             repo,
             ctx1a,
             ctx1b,
             ctx2,
             matcher,
             tmproot,
             cmdline,
             do3way,
             guitool,
             opts,
         )
 
     finally:
         ui.note(_(b'cleaning up temp directory\n'))
         shutil.rmtree(tmproot)
 
 
 extdiffopts = (
     [
         (
             b'o',
             b'option',
             [],
             _(b'pass option to comparison program'),
             _(b'OPT'),
         ),
         (b'r', b'rev', [], _(b'revision (DEPRECATED)'), _(b'REV')),
         (b'', b'from', b'', _(b'revision to diff from'), _(b'REV1')),
         (b'', b'to', b'', _(b'revision to diff to'), _(b'REV2')),
         (b'c', b'change', b'', _(b'change made by revision'), _(b'REV')),
         (
             b'',
             b'per-file',
             False,
             _(b'compare each file instead of revision snapshots'),
         ),
         (
             b'',
             b'confirm',
             False,
             _(b'prompt user before each external program invocation'),
         ),
         (b'', b'patch', None, _(b'compare patches for two revisions')),
     ]
     + cmdutil.walkopts
     + cmdutil.subrepoopts
 )
 
 
 @command(
     b'extdiff',
     [
         (b'p', b'program', b'', _(b'comparison program to run'), _(b'CMD')),
     ]
     + extdiffopts,
     _(b'hg extdiff [OPT]... [FILE]...'),
     helpcategory=command.CATEGORY_FILE_CONTENTS,
     inferrepo=True,
 )
 def extdiff(ui, repo, *pats, **opts):
     """use external program to diff repository (or selected files)
 
     Show differences between revisions for the specified files, using
     an external program. The default program used is diff, with
     default options "-Npru".
 
     To select a different program, use the -p/--program option. The
     program will be passed the names of two directories to compare,
     unless the --per-file option is specified (see below). To pass
     additional options to the program, use -o/--option. These will be
     passed before the names of the directories or files to compare.
 
     The --from, --to, and --change options work the same way they do for
     :hg:`diff`.
 
     The --per-file option runs the external program repeatedly on each
     file to diff, instead of once on two directories. By default,
     this happens one by one, where the next file diff is open in the
     external program only once the previous external program (for the
     previous file diff) has exited. If the external program has a
     graphical interface, it can open all the file diffs at once instead
     of one by one. See :hg:`help -e extdiff` for information about how
     to tell Mercurial that a given program has a graphical interface.
 
     The --confirm option will prompt the user before each invocation of
     the external program. It is ignored if --per-file isn't specified.
     """
     opts = pycompat.byteskwargs(opts)
     program = opts.get(b'program')
     option = opts.get(b'option')
     if not program:
         program = b'diff'
         option = option or [b'-Npru']
     cmdline = b' '.join(map(procutil.shellquote, [program] + option))
     return dodiff(ui, repo, cmdline, pats, opts)
 
 
 class savedcmd(object):
     """use external program to diff repository (or selected files)
 
     Show differences between revisions for the specified files, using
     the following program::
 
         %(path)s
 
     When two revision arguments are given, then changes are shown
     between those revisions. If only one revision is specified then
     that revision is compared to the working directory, and, when no
     revisions are specified, the working directory files are compared
     to its parent.
     """
 
     def __init__(self, path, cmdline, isgui):
         # We can't pass non-ASCII through docstrings (and path is
         # in an unknown encoding anyway), but avoid double separators on
         # Windows
         docpath = stringutil.escapestr(path).replace(b'\\\\', b'\\')
         self.__doc__ %= {'path': pycompat.sysstr(stringutil.uirepr(docpath))}
         self._cmdline = cmdline
         self._isgui = isgui
 
     def __call__(self, ui, repo, *pats, **opts):
         opts = pycompat.byteskwargs(opts)
         options = b' '.join(map(procutil.shellquote, opts[b'option']))
         if options:
             options = b' ' + options
         return dodiff(
             ui, repo, self._cmdline + options, pats, opts, guitool=self._isgui
         )
 
 
 def _gettooldetails(ui, cmd, path):
     """
     returns following things for a
     ```
     [extdiff]
     <cmd> = <path>
     ```
     entry:
 
     cmd: command/tool name
     path: path to the tool
     cmdline: the command which should be run
     isgui: whether the tool uses GUI or not
 
     Reads all external tools related configs, whether it be extdiff section,
     diff-tools or merge-tools section, or its specified in an old format or
     the latest format.
     """
     path = util.expandpath(path)
     if cmd.startswith(b'cmd.'):
         cmd = cmd[4:]
         if not path:
             path = procutil.findexe(cmd)
             if path is None:
                 path = filemerge.findexternaltool(ui, cmd) or cmd
         diffopts = ui.config(b'extdiff', b'opts.' + cmd)
         cmdline = procutil.shellquote(path)
         if diffopts:
             cmdline += b' ' + diffopts
         isgui = ui.configbool(b'extdiff', b'gui.' + cmd)
     else:
         if path:
             # case "cmd = path opts"
             cmdline = path
             diffopts = len(pycompat.shlexsplit(cmdline)) > 1
         else:
             # case "cmd ="
             path = procutil.findexe(cmd)
             if path is None:
                 path = filemerge.findexternaltool(ui, cmd) or cmd
             cmdline = procutil.shellquote(path)
             diffopts = False
         isgui = ui.configbool(b'extdiff', b'gui.' + cmd)
     # look for diff arguments in [diff-tools] then [merge-tools]
     if not diffopts:
         key = cmd + b'.diffargs'
         for section in (b'diff-tools', b'merge-tools'):
             args = ui.config(section, key)
             if args:
                 cmdline += b' ' + args
                 if isgui is None:
                     isgui = ui.configbool(section, cmd + b'.gui') or False
                 break
     return cmd, path, cmdline, isgui
 
 
 def uisetup(ui):
     for cmd, path in ui.configitems(b'extdiff'):
         if cmd.startswith(b'opts.') or cmd.startswith(b'gui.'):
             continue
         cmd, path, cmdline, isgui = _gettooldetails(ui, cmd, path)
         command(
             cmd,
             extdiffopts[:],
             _(b'hg %s [OPTION]... [FILE]...') % cmd,
             helpcategory=command.CATEGORY_FILE_CONTENTS,
             inferrepo=True,
         )(savedcmd(path, cmdline, isgui))
 
 
 # tell hggettext to extract docstrings from these functions:
 i18nfunctions = [savedcmd]

hgext/split.py

@@ -1,197 +1,197 b''
 # split.py - split a changeset into smaller ones
 #
 # Copyright 2015 Laurent Charignon <[e-mail]>
 # Copyright 2017 Facebook, Inc.
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 """command to split a changeset into smaller ones (EXPERIMENTAL)"""
 
 from __future__ import absolute_import
 
 from mercurial.i18n import _
 
 from mercurial.node import (
-    nullid,
+    nullrev,
     short,
 )
 
 from mercurial import (
     bookmarks,
     cmdutil,
     commands,
     error,
     hg,
     pycompat,
     registrar,
     revsetlang,
     rewriteutil,
     scmutil,
     util,
 )
 
 # allow people to use split without explicitly enabling rebase extension
 from . import rebase
 
 cmdtable = {}
 command = registrar.command(cmdtable)
 
 # Note for extension authors: ONLY specify testedwith = 'ships-with-hg-core' for
 # extensions which SHIP WITH MERCURIAL. Non-mainline extensions should
 # be specifying the version(s) of Mercurial they are tested with, or
 # leave the attribute unspecified.
 testedwith = b'ships-with-hg-core'
 
 
 @command(
     b'split',
     [
         (b'r', b'rev', b'', _(b"revision to split"), _(b'REV')),
         (b'', b'rebase', True, _(b'rebase descendants after split')),
     ]
     + cmdutil.commitopts2,
     _(b'hg split [--no-rebase] [[-r] REV]'),
     helpcategory=command.CATEGORY_CHANGE_MANAGEMENT,
     helpbasic=True,
 )
 def split(ui, repo, *revs, **opts):
     """split a changeset into smaller ones
 
     Repeatedly prompt changes and commit message for new changesets until there
     is nothing left in the original changeset.
 
     If --rev was not given, split the working directory parent.
 
     By default, rebase connected non-obsoleted descendants onto the new
     changeset. Use --no-rebase to avoid the rebase.
     """
     opts = pycompat.byteskwargs(opts)
     revlist = []
     if opts.get(b'rev'):
         revlist.append(opts.get(b'rev'))
     revlist.extend(revs)
     with repo.wlock(), repo.lock():
         tr = repo.transaction(b'split')
         # If the rebase somehow runs into conflicts, make sure
         # we close the transaction so the user can continue it.
         with util.acceptintervention(tr):
             revs = scmutil.revrange(repo, revlist or [b'.'])
             if len(revs) > 1:
                 raise error.InputError(_(b'cannot split multiple revisions'))
 
             rev = revs.first()
-            ctx = repo[rev]
-            # Handle nullid specially here (instead of leaving for precheck()
+            # Handle nullrev specially here (instead of leaving for precheck()
             # below) so we get a nicer message and error code.
-            if rev is None or ctx.node() == nullid:
+            if rev is None or rev == nullrev:
                 ui.status(_(b'nothing to split\n'))
                 return 1
+            ctx = repo[rev]
             if ctx.node() is None:
                 raise error.InputError(_(b'cannot split working directory'))
 
             if opts.get(b'rebase'):
                 # Skip obsoleted descendants and their descendants so the rebase
                 # won't cause conflicts for sure.
                 descendants = list(repo.revs(b'(%d::) - (%d)', rev, rev))
                 torebase = list(
                     repo.revs(
                         b'%ld - (%ld & obsolete())::', descendants, descendants
                     )
                 )
             else:
                 torebase = []
             rewriteutil.precheck(repo, [rev] + torebase, b'split')
 
             if len(ctx.parents()) > 1:
                 raise error.InputError(_(b'cannot split a merge changeset'))
 
             cmdutil.bailifchanged(repo)
 
             # Deactivate bookmark temporarily so it won't get moved
             # unintentionally
             bname = repo._activebookmark
             if bname and repo._bookmarks[bname] != ctx.node():
                 bookmarks.deactivate(repo)
 
             wnode = repo[b'.'].node()
             top = None
             try:
                 top = dosplit(ui, repo, tr, ctx, opts)
             finally:
                 # top is None: split failed, need update --clean recovery.
                 # wnode == ctx.node(): wnode split, no need to update.
                 if top is None or wnode != ctx.node():
                     hg.clean(repo, wnode, show_stats=False)
                 if bname:
                     bookmarks.activate(repo, bname)
             if torebase and top:
                 dorebase(ui, repo, torebase, top)
 
 
 def dosplit(ui, repo, tr, ctx, opts):
     committed = []  # [ctx]
 
     # Set working parent to ctx.p1(), and keep working copy as ctx's content
     if ctx.node() != repo.dirstate.p1():
         hg.clean(repo, ctx.node(), show_stats=False)
     with repo.dirstate.parentchange():
         scmutil.movedirstate(repo, ctx.p1())
 
     # Any modified, added, removed, deleted result means split is incomplete
     def incomplete(repo):
         st = repo.status()
         return any((st.modified, st.added, st.removed, st.deleted))
 
     # Main split loop
     while incomplete(repo):
         if committed:
             header = _(
                 b'HG: Splitting %s. So far it has been split into:\n'
             ) % short(ctx.node())
             # We don't want color codes in the commit message template, so
             # disable the label() template function while we render it.
             with ui.configoverride(
                 {(b'templatealias', b'label(l,x)'): b"x"}, b'split'
             ):
                 for c in committed:
                     summary = cmdutil.format_changeset_summary(ui, c, b'split')
                     header += _(b'HG: - %s\n') % summary
             header += _(
                 b'HG: Write commit message for the next split changeset.\n'
             )
         else:
             header = _(
                 b'HG: Splitting %s. Write commit message for the '
                 b'first split changeset.\n'
             ) % short(ctx.node())
         opts.update(
             {
                 b'edit': True,
                 b'interactive': True,
                 b'message': header + ctx.description(),
             }
         )
         commands.commit(ui, repo, **pycompat.strkwargs(opts))
         newctx = repo[b'.']
         committed.append(newctx)
 
     if not committed:
         raise error.InputError(_(b'cannot split an empty revision'))
 
     scmutil.cleanupnodes(
         repo,
         {ctx.node(): [c.node() for c in committed]},
         operation=b'split',
         fixphase=True,
     )
 
     return committed[-1]
 
 
 def dorebase(ui, repo, src, destctx):
     rebase.rebase(
         ui,
         repo,
         rev=[revsetlang.formatspec(b'%ld', src)],
         dest=revsetlang.formatspec(b'%d', destctx.rev()),
     )

mercurial/context.py

@@ -1,3113 +1,3113 b''
 # context.py - changeset and file context objects for mercurial
 #
 # Copyright 2006, 2007 Olivia Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import errno
 import filecmp
 import os
 import stat
 
 from .i18n import _
 from .node import (
     addednodeid,
     hex,
     modifiednodeid,
     nullid,
     nullrev,
     short,
     wdirfilenodeids,
     wdirhex,
 )
 from .pycompat import (
     getattr,
     open,
 )
 from . import (
     dagop,
     encoding,
     error,
     fileset,
     match as matchmod,
     mergestate as mergestatemod,
     metadata,
     obsolete as obsmod,
     patch,
     pathutil,
     phases,
     pycompat,
     repoview,
     scmutil,
     sparse,
     subrepo,
     subrepoutil,
     util,
 )
 from .utils import (
     dateutil,
     stringutil,
 )
 
 propertycache = util.propertycache
 
 
 class basectx(object):
     """A basectx object represents the common logic for its children:
     changectx: read-only context that is already present in the repo,
     workingctx: a context that represents the working directory and can
                 be committed,
     memctx: a context that represents changes in-memory and can also
             be committed."""
 
     def __init__(self, repo):
         self._repo = repo
 
     def __bytes__(self):
         return short(self.node())
 
     __str__ = encoding.strmethod(__bytes__)
 
     def __repr__(self):
         return "<%s %s>" % (type(self).__name__, str(self))
 
     def __eq__(self, other):
         try:
             return type(self) == type(other) and self._rev == other._rev
         except AttributeError:
             return False
 
     def __ne__(self, other):
         return not (self == other)
 
     def __contains__(self, key):
         return key in self._manifest
 
     def __getitem__(self, key):
         return self.filectx(key)
 
     def __iter__(self):
         return iter(self._manifest)
 
     def _buildstatusmanifest(self, status):
         """Builds a manifest that includes the given status results, if this is
         a working copy context. For non-working copy contexts, it just returns
         the normal manifest."""
         return self.manifest()
 
     def _matchstatus(self, other, match):
         """This internal method provides a way for child objects to override the
         match operator.
         """
         return match
 
     def _buildstatus(
         self, other, s, match, listignored, listclean, listunknown
     ):
         """build a status with respect to another context"""
         # Load earliest manifest first for caching reasons. More specifically,
         # if you have revisions 1000 and 1001, 1001 is probably stored as a
         # delta against 1000. Thus, if you read 1000 first, we'll reconstruct
         # 1000 and cache it so that when you read 1001, we just need to apply a
         # delta to what's in the cache. So that's one full reconstruction + one
         # delta application.
         mf2 = None
         if self.rev() is not None and self.rev() < other.rev():
             mf2 = self._buildstatusmanifest(s)
         mf1 = other._buildstatusmanifest(s)
         if mf2 is None:
             mf2 = self._buildstatusmanifest(s)
 
         modified, added = [], []
         removed = []
         clean = []
         deleted, unknown, ignored = s.deleted, s.unknown, s.ignored
         deletedset = set(deleted)
         d = mf1.diff(mf2, match=match, clean=listclean)
         for fn, value in pycompat.iteritems(d):
             if fn in deletedset:
                 continue
             if value is None:
                 clean.append(fn)
                 continue
             (node1, flag1), (node2, flag2) = value
             if node1 is None:
                 added.append(fn)
             elif node2 is None:
                 removed.append(fn)
             elif flag1 != flag2:
                 modified.append(fn)
             elif node2 not in wdirfilenodeids:
                 # When comparing files between two commits, we save time by
                 # not comparing the file contents when the nodeids differ.
                 # Note that this means we incorrectly report a reverted change
                 # to a file as a modification.
                 modified.append(fn)
             elif self[fn].cmp(other[fn]):
                 modified.append(fn)
             else:
                 clean.append(fn)
 
         if removed:
             # need to filter files if they are already reported as removed
             unknown = [
                 fn
                 for fn in unknown
                 if fn not in mf1 and (not match or match(fn))
             ]
             ignored = [
                 fn
                 for fn in ignored
                 if fn not in mf1 and (not match or match(fn))
             ]
             # if they're deleted, don't report them as removed
             removed = [fn for fn in removed if fn not in deletedset]
 
         return scmutil.status(
             modified, added, removed, deleted, unknown, ignored, clean
         )
 
     @propertycache
     def substate(self):
         return subrepoutil.state(self, self._repo.ui)
 
     def subrev(self, subpath):
         return self.substate[subpath][1]
 
     def rev(self):
         return self._rev
 
     def node(self):
         return self._node
 
     def hex(self):
         return hex(self.node())
 
     def manifest(self):
         return self._manifest
 
     def manifestctx(self):
         return self._manifestctx
 
     def repo(self):
         return self._repo
 
     def phasestr(self):
         return phases.phasenames[self.phase()]
 
     def mutable(self):
         return self.phase() > phases.public
 
     def matchfileset(self, cwd, expr, badfn=None):
         return fileset.match(self, cwd, expr, badfn=badfn)
 
     def obsolete(self):
         """True if the changeset is obsolete"""
         return self.rev() in obsmod.getrevs(self._repo, b'obsolete')
 
     def extinct(self):
         """True if the changeset is extinct"""
         return self.rev() in obsmod.getrevs(self._repo, b'extinct')
 
     def orphan(self):
         """True if the changeset is not obsolete, but its ancestor is"""
         return self.rev() in obsmod.getrevs(self._repo, b'orphan')
 
     def phasedivergent(self):
         """True if the changeset tries to be a successor of a public changeset
 
         Only non-public and non-obsolete changesets may be phase-divergent.
         """
         return self.rev() in obsmod.getrevs(self._repo, b'phasedivergent')
 
     def contentdivergent(self):
         """Is a successor of a changeset with multiple possible successor sets
 
         Only non-public and non-obsolete changesets may be content-divergent.
         """
         return self.rev() in obsmod.getrevs(self._repo, b'contentdivergent')
 
     def isunstable(self):
         """True if the changeset is either orphan, phase-divergent or
         content-divergent"""
         return self.orphan() or self.phasedivergent() or self.contentdivergent()
 
     def instabilities(self):
         """return the list of instabilities affecting this changeset.
 
         Instabilities are returned as strings. possible values are:
         - orphan,
         - phase-divergent,
         - content-divergent.
         """
         instabilities = []
         if self.orphan():
             instabilities.append(b'orphan')
         if self.phasedivergent():
             instabilities.append(b'phase-divergent')
         if self.contentdivergent():
             instabilities.append(b'content-divergent')
         return instabilities
 
     def parents(self):
         """return contexts for each parent changeset"""
         return self._parents
 
     def p1(self):
         return self._parents[0]
 
     def p2(self):
         parents = self._parents
         if len(parents) == 2:
             return parents[1]
         return self._repo[nullrev]
 
     def _fileinfo(self, path):
         if '_manifest' in self.__dict__:
             try:
                 return self._manifest.find(path)
             except KeyError:
                 raise error.ManifestLookupError(
                     self._node or b'None', path, _(b'not found in manifest')
                 )
         if '_manifestdelta' in self.__dict__ or path in self.files():
             if path in self._manifestdelta:
                 return (
                     self._manifestdelta[path],
                     self._manifestdelta.flags(path),
                 )
         mfl = self._repo.manifestlog
         try:
             node, flag = mfl[self._changeset.manifest].find(path)
         except KeyError:
             raise error.ManifestLookupError(
                 self._node or b'None', path, _(b'not found in manifest')
             )
 
         return node, flag
 
     def filenode(self, path):
         return self._fileinfo(path)[0]
 
     def flags(self, path):
         try:
             return self._fileinfo(path)[1]
         except error.LookupError:
             return b''
 
     @propertycache
     def _copies(self):
         return metadata.computechangesetcopies(self)
 
     def p1copies(self):
         return self._copies[0]
 
     def p2copies(self):
         return self._copies[1]
 
     def sub(self, path, allowcreate=True):
         '''return a subrepo for the stored revision of path, never wdir()'''
         return subrepo.subrepo(self, path, allowcreate=allowcreate)
 
     def nullsub(self, path, pctx):
         return subrepo.nullsubrepo(self, path, pctx)
 
     def workingsub(self, path):
         """return a subrepo for the stored revision, or wdir if this is a wdir
         context.
         """
         return subrepo.subrepo(self, path, allowwdir=True)
 
     def match(
         self,
         pats=None,
         include=None,
         exclude=None,
         default=b'glob',
         listsubrepos=False,
         badfn=None,
         cwd=None,
     ):
         r = self._repo
         if not cwd:
             cwd = r.getcwd()
         return matchmod.match(
             r.root,
             cwd,
             pats,
             include,
             exclude,
             default,
             auditor=r.nofsauditor,
             ctx=self,
             listsubrepos=listsubrepos,
             badfn=badfn,
         )
 
     def diff(
         self,
         ctx2=None,
         match=None,
         changes=None,
         opts=None,
         losedatafn=None,
         pathfn=None,
         copy=None,
         copysourcematch=None,
         hunksfilterfn=None,
     ):
         """Returns a diff generator for the given contexts and matcher"""
         if ctx2 is None:
             ctx2 = self.p1()
         if ctx2 is not None:
             ctx2 = self._repo[ctx2]
         return patch.diff(
             self._repo,
             ctx2,
             self,
             match=match,
             changes=changes,
             opts=opts,
             losedatafn=losedatafn,
             pathfn=pathfn,
             copy=copy,
             copysourcematch=copysourcematch,
             hunksfilterfn=hunksfilterfn,
         )
 
     def dirs(self):
         return self._manifest.dirs()
 
     def hasdir(self, dir):
         return self._manifest.hasdir(dir)
 
     def status(
         self,
         other=None,
         match=None,
         listignored=False,
         listclean=False,
         listunknown=False,
         listsubrepos=False,
     ):
         """return status of files between two nodes or node and working
         directory.
 
         If other is None, compare this node with working directory.
 
         ctx1.status(ctx2) returns the status of change from ctx1 to ctx2
 
         Returns a mercurial.scmutils.status object.
 
         Data can be accessed using either tuple notation:
 
             (modified, added, removed, deleted, unknown, ignored, clean)
 
         or direct attribute access:
 
             s.modified, s.added, ...
         """
 
         ctx1 = self
         ctx2 = self._repo[other]
 
         # This next code block is, admittedly, fragile logic that tests for
         # reversing the contexts and wouldn't need to exist if it weren't for
         # the fast (and common) code path of comparing the working directory
         # with its first parent.
         #
         # What we're aiming for here is the ability to call:
         #
         # workingctx.status(parentctx)
         #
         # If we always built the manifest for each context and compared those,
         # then we'd be done. But the special case of the above call means we
         # just copy the manifest of the parent.
         reversed = False
         if not isinstance(ctx1, changectx) and isinstance(ctx2, changectx):
             reversed = True
             ctx1, ctx2 = ctx2, ctx1
 
         match = self._repo.narrowmatch(match)
         match = ctx2._matchstatus(ctx1, match)
         r = scmutil.status([], [], [], [], [], [], [])
         r = ctx2._buildstatus(
             ctx1, r, match, listignored, listclean, listunknown
         )
 
         if reversed:
             # Reverse added and removed. Clear deleted, unknown and ignored as
             # these make no sense to reverse.
             r = scmutil.status(
                 r.modified, r.removed, r.added, [], [], [], r.clean
             )
 
         if listsubrepos:
             for subpath, sub in scmutil.itersubrepos(ctx1, ctx2):
                 try:
                     rev2 = ctx2.subrev(subpath)
                 except KeyError:
                     # A subrepo that existed in node1 was deleted between
                     # node1 and node2 (inclusive). Thus, ctx2's substate
                     # won't contain that subpath. The best we can do ignore it.
                     rev2 = None
                 submatch = matchmod.subdirmatcher(subpath, match)
                 s = sub.status(
                     rev2,
                     match=submatch,
                     ignored=listignored,
                     clean=listclean,
                     unknown=listunknown,
                     listsubrepos=True,
                 )
                 for k in (
                     'modified',
                     'added',
                     'removed',
                     'deleted',
                     'unknown',
                     'ignored',
                     'clean',
                 ):
                     rfiles, sfiles = getattr(r, k), getattr(s, k)
                     rfiles.extend(b"%s/%s" % (subpath, f) for f in sfiles)
 
         r.modified.sort()
         r.added.sort()
         r.removed.sort()
         r.deleted.sort()
         r.unknown.sort()
         r.ignored.sort()
         r.clean.sort()
 
         return r
 
     def mergestate(self, clean=False):
         """Get a mergestate object for this context."""
         raise NotImplementedError(
             '%s does not implement mergestate()' % self.__class__
         )
 
     def isempty(self):
         return not (
             len(self.parents()) > 1
             or self.branch() != self.p1().branch()
             or self.closesbranch()
             or self.files()
         )
 
 
 class changectx(basectx):
     """A changecontext object makes access to data related to a particular
     changeset convenient. It represents a read-only context already present in
     the repo."""
 
     def __init__(self, repo, rev, node, maybe_filtered=True):
         super(changectx, self).__init__(repo)
         self._rev = rev
         self._node = node
         # When maybe_filtered is True, the revision might be affected by
         # changelog filtering and operation through the filtered changelog must be used.
         #
         # When maybe_filtered is False, the revision has already been checked
         # against filtering and is not filtered. Operation through the
         # unfiltered changelog might be used in some case.
         self._maybe_filtered = maybe_filtered
 
     def __hash__(self):
         try:
             return hash(self._rev)
         except AttributeError:
             return id(self)
 
     def __nonzero__(self):
         return self._rev != nullrev
 
     __bool__ = __nonzero__
 
     @propertycache
     def _changeset(self):
         if self._maybe_filtered:
             repo = self._repo
         else:
             repo = self._repo.unfiltered()
         return repo.changelog.changelogrevision(self.rev())
 
     @propertycache
     def _manifest(self):
         return self._manifestctx.read()
 
     @property
     def _manifestctx(self):
         return self._repo.manifestlog[self._changeset.manifest]
 
     @propertycache
     def _manifestdelta(self):
         return self._manifestctx.readdelta()
 
     @propertycache
     def _parents(self):
         repo = self._repo
         if self._maybe_filtered:
             cl = repo.changelog
         else:
             cl = repo.unfiltered().changelog
 
         p1, p2 = cl.parentrevs(self._rev)
         if p2 == nullrev:
             return [changectx(repo, p1, cl.node(p1), maybe_filtered=False)]
         return [
             changectx(repo, p1, cl.node(p1), maybe_filtered=False),
             changectx(repo, p2, cl.node(p2), maybe_filtered=False),
         ]
 
     def changeset(self):
         c = self._changeset
         return (
             c.manifest,
             c.user,
             c.date,
             c.files,
             c.description,
             c.extra,
         )
 
     def manifestnode(self):
         return self._changeset.manifest
 
     def user(self):
         return self._changeset.user
 
     def date(self):
         return self._changeset.date
 
     def files(self):
         return self._changeset.files
 
     def filesmodified(self):
         modified = set(self.files())
         modified.difference_update(self.filesadded())
         modified.difference_update(self.filesremoved())
         return sorted(modified)
 
     def filesadded(self):
         filesadded = self._changeset.filesadded
         compute_on_none = True
         if self._repo.filecopiesmode == b'changeset-sidedata':
             compute_on_none = False
         else:
             source = self._repo.ui.config(b'experimental', b'copies.read-from')
             if source == b'changeset-only':
                 compute_on_none = False
             elif source != b'compatibility':
                 # filelog mode, ignore any changelog content
                 filesadded = None
         if filesadded is None:
             if compute_on_none:
                 filesadded = metadata.computechangesetfilesadded(self)
             else:
                 filesadded = []
         return filesadded
 
     def filesremoved(self):
         filesremoved = self._changeset.filesremoved
         compute_on_none = True
         if self._repo.filecopiesmode == b'changeset-sidedata':
             compute_on_none = False
         else:
             source = self._repo.ui.config(b'experimental', b'copies.read-from')
             if source == b'changeset-only':
                 compute_on_none = False
             elif source != b'compatibility':
                 # filelog mode, ignore any changelog content
                 filesremoved = None
         if filesremoved is None:
             if compute_on_none:
                 filesremoved = metadata.computechangesetfilesremoved(self)
             else:
                 filesremoved = []
         return filesremoved
 
     @propertycache
     def _copies(self):
         p1copies = self._changeset.p1copies
         p2copies = self._changeset.p2copies
         compute_on_none = True
         if self._repo.filecopiesmode == b'changeset-sidedata':
             compute_on_none = False
         else:
             source = self._repo.ui.config(b'experimental', b'copies.read-from')
             # If config says to get copy metadata only from changeset, then
             # return that, defaulting to {} if there was no copy metadata.  In
             # compatibility mode, we return copy data from the changeset if it
             # was recorded there, and otherwise we fall back to getting it from
             # the filelogs (below).
             #
             # If we are in compatiblity mode and there is not data in the
             # changeset), we get the copy metadata from the filelogs.
             #
             # otherwise, when config said to read only from filelog, we get the
             # copy metadata from the filelogs.
             if source == b'changeset-only':
                 compute_on_none = False
             elif source != b'compatibility':
                 # filelog mode, ignore any changelog content
                 p1copies = p2copies = None
         if p1copies is None:
             if compute_on_none:
                 p1copies, p2copies = super(changectx, self)._copies
             else:
                 if p1copies is None:
                     p1copies = {}
         if p2copies is None:
             p2copies = {}
         return p1copies, p2copies
 
     def description(self):
         return self._changeset.description
 
     def branch(self):
         return encoding.tolocal(self._changeset.extra.get(b"branch"))
 
     def closesbranch(self):
         return b'close' in self._changeset.extra
 
     def extra(self):
         """Return a dict of extra information."""
         return self._changeset.extra
 
     def tags(self):
         """Return a list of byte tag names"""
         return self._repo.nodetags(self._node)
 
     def bookmarks(self):
         """Return a list of byte bookmark names."""
         return self._repo.nodebookmarks(self._node)
 
     def phase(self):
         return self._repo._phasecache.phase(self._repo, self._rev)
 
     def hidden(self):
         return self._rev in repoview.filterrevs(self._repo, b'visible')
 
     def isinmemory(self):
         return False
 
     def children(self):
         """return list of changectx contexts for each child changeset.
 
         This returns only the immediate child changesets. Use descendants() to
         recursively walk children.
         """
         c = self._repo.changelog.children(self._node)
         return [self._repo[x] for x in c]
 
     def ancestors(self):
         for a in self._repo.changelog.ancestors([self._rev]):
             yield self._repo[a]
 
     def descendants(self):
         """Recursively yield all children of the changeset.
 
         For just the immediate children, use children()
         """
         for d in self._repo.changelog.descendants([self._rev]):
             yield self._repo[d]
 
     def filectx(self, path, fileid=None, filelog=None):
         """get a file context from this changeset"""
         if fileid is None:
             fileid = self.filenode(path)
         return filectx(
             self._repo, path, fileid=fileid, changectx=self, filelog=filelog
         )
 
     def ancestor(self, c2, warn=False):
         """return the "best" ancestor context of self and c2
 
         If there are multiple candidates, it will show a message and check
         merge.preferancestor configuration before falling back to the
         revlog ancestor."""
         # deal with workingctxs
         n2 = c2._node
         if n2 is None:
             n2 = c2._parents[0]._node
         cahs = self._repo.changelog.commonancestorsheads(self._node, n2)
         if not cahs:
             anc = nullid
         elif len(cahs) == 1:
             anc = cahs[0]
         else:
             # experimental config: merge.preferancestor
             for r in self._repo.ui.configlist(b'merge', b'preferancestor'):
                 try:
                     ctx = scmutil.revsymbol(self._repo, r)
                 except error.RepoLookupError:
                     continue
                 anc = ctx.node()
                 if anc in cahs:
                     break
             else:
                 anc = self._repo.changelog.ancestor(self._node, n2)
             if warn:
                 self._repo.ui.status(
                     (
                         _(b"note: using %s as ancestor of %s and %s\n")
                         % (short(anc), short(self._node), short(n2))
                     )
                     + b''.join(
                         _(
                             b"      alternatively, use --config "
                             b"merge.preferancestor=%s\n"
                         )
                         % short(n)
                         for n in sorted(cahs)
                         if n != anc
                     )
                 )
         return self._repo[anc]
 
     def isancestorof(self, other):
         """True if this changeset is an ancestor of other"""
         return self._repo.changelog.isancestorrev(self._rev, other._rev)
 
     def walk(self, match):
         '''Generates matching file names.'''
 
         # Wrap match.bad method to have message with nodeid
         def bad(fn, msg):
             # The manifest doesn't know about subrepos, so don't complain about
             # paths into valid subrepos.
             if any(fn == s or fn.startswith(s + b'/') for s in self.substate):
                 return
             match.bad(fn, _(b'no such file in rev %s') % self)
 
         m = matchmod.badmatch(self._repo.narrowmatch(match), bad)
         return self._manifest.walk(m)
 
     def matches(self, match):
         return self.walk(match)
 
 
 class basefilectx(object):
     """A filecontext object represents the common logic for its children:
     filectx: read-only access to a filerevision that is already present
              in the repo,
     workingfilectx: a filecontext that represents files from the working
                     directory,
     memfilectx: a filecontext that represents files in-memory,
     """
 
     @propertycache
     def _filelog(self):
         return self._repo.file(self._path)
 
     @propertycache
     def _changeid(self):
         if '_changectx' in self.__dict__:
             return self._changectx.rev()
         elif '_descendantrev' in self.__dict__:
             # this file context was created from a revision with a known
             # descendant, we can (lazily) correct for linkrev aliases
             return self._adjustlinkrev(self._descendantrev)
         else:
             return self._filelog.linkrev(self._filerev)
 
     @propertycache
     def _filenode(self):
         if '_fileid' in self.__dict__:
             return self._filelog.lookup(self._fileid)
         else:
             return self._changectx.filenode(self._path)
 
     @propertycache
     def _filerev(self):
         return self._filelog.rev(self._filenode)
 
     @propertycache
     def _repopath(self):
         return self._path
 
     def __nonzero__(self):
         try:
             self._filenode
             return True
         except error.LookupError:
             # file is missing
             return False
 
     __bool__ = __nonzero__
 
     def __bytes__(self):
         try:
             return b"%s@%s" % (self.path(), self._changectx)
         except error.LookupError:
             return b"%s@???" % self.path()
 
     __str__ = encoding.strmethod(__bytes__)
 
     def __repr__(self):
         return "<%s %s>" % (type(self).__name__, str(self))
 
     def __hash__(self):
         try:
             return hash((self._path, self._filenode))
         except AttributeError:
             return id(self)
 
     def __eq__(self, other):
         try:
             return (
                 type(self) == type(other)
                 and self._path == other._path
                 and self._filenode == other._filenode
             )
         except AttributeError:
             return False
 
     def __ne__(self, other):
         return not (self == other)
 
     def filerev(self):
         return self._filerev
 
     def filenode(self):
         return self._filenode
 
     @propertycache
     def _flags(self):
         return self._changectx.flags(self._path)
 
     def flags(self):
         return self._flags
 
     def filelog(self):
         return self._filelog
 
     def rev(self):
         return self._changeid
 
     def linkrev(self):
         return self._filelog.linkrev(self._filerev)
 
     def node(self):
         return self._changectx.node()
 
     def hex(self):
         return self._changectx.hex()
 
     def user(self):
         return self._changectx.user()
 
     def date(self):
         return self._changectx.date()
 
     def files(self):
         return self._changectx.files()
 
     def description(self):
         return self._changectx.description()
 
     def branch(self):
         return self._changectx.branch()
 
     def extra(self):
         return self._changectx.extra()
 
     def phase(self):
         return self._changectx.phase()
 
     def phasestr(self):
         return self._changectx.phasestr()
 
     def obsolete(self):
         return self._changectx.obsolete()
 
     def instabilities(self):
         return self._changectx.instabilities()
 
     def manifest(self):
         return self._changectx.manifest()
 
     def changectx(self):
         return self._changectx
 
     def renamed(self):
         return self._copied
 
     def copysource(self):
         return self._copied and self._copied[0]
 
     def repo(self):
         return self._repo
 
     def size(self):
         return len(self.data())
 
     def path(self):
         return self._path
 
     def isbinary(self):
         try:
             return stringutil.binary(self.data())
         except IOError:
             return False
 
     def isexec(self):
         return b'x' in self.flags()
 
     def islink(self):
         return b'l' in self.flags()
 
     def isabsent(self):
         """whether this filectx represents a file not in self._changectx
 
         This is mainly for merge code to detect change/delete conflicts. This is
         expected to be True for all subclasses of basectx."""
         return False
 
     _customcmp = False
 
     def cmp(self, fctx):
         """compare with other file context
 
         returns True if different than fctx.
         """
         if fctx._customcmp:
             return fctx.cmp(self)
 
         if self._filenode is None:
             raise error.ProgrammingError(
                 b'filectx.cmp() must be reimplemented if not backed by revlog'
             )
 
         if fctx._filenode is None:
             if self._repo._encodefilterpats:
                 # can't rely on size() because wdir content may be decoded
                 return self._filelog.cmp(self._filenode, fctx.data())
             if self.size() - 4 == fctx.size():
                 # size() can match:
                 # if file data starts with '\1\n', empty metadata block is
                 # prepended, which adds 4 bytes to filelog.size().
                 return self._filelog.cmp(self._filenode, fctx.data())
         if self.size() == fctx.size() or self.flags() == b'l':
             # size() matches: need to compare content
             # issue6456: Always compare symlinks because size can represent
             # encrypted string for EXT-4 encryption(fscrypt).
             return self._filelog.cmp(self._filenode, fctx.data())
 
         # size() differs
         return True
 
     def _adjustlinkrev(self, srcrev, inclusive=False, stoprev=None):
         """return the first ancestor of <srcrev> introducing <fnode>
 
         If the linkrev of the file revision does not point to an ancestor of
         srcrev, we'll walk down the ancestors until we find one introducing
         this file revision.
 
         :srcrev: the changeset revision we search ancestors from
         :inclusive: if true, the src revision will also be checked
         :stoprev: an optional revision to stop the walk at. If no introduction
                   of this file content could be found before this floor
                   revision, the function will returns "None" and stops its
                   iteration.
         """
         repo = self._repo
         cl = repo.unfiltered().changelog
         mfl = repo.manifestlog
         # fetch the linkrev
         lkr = self.linkrev()
         if srcrev == lkr:
             return lkr
         # hack to reuse ancestor computation when searching for renames
         memberanc = getattr(self, '_ancestrycontext', None)
         iteranc = None
         if srcrev is None:
             # wctx case, used by workingfilectx during mergecopy
             revs = [p.rev() for p in self._repo[None].parents()]
             inclusive = True  # we skipped the real (revless) source
         else:
             revs = [srcrev]
         if memberanc is None:
             memberanc = iteranc = cl.ancestors(revs, lkr, inclusive=inclusive)
         # check if this linkrev is an ancestor of srcrev
         if lkr not in memberanc:
             if iteranc is None:
                 iteranc = cl.ancestors(revs, lkr, inclusive=inclusive)
             fnode = self._filenode
             path = self._path
             for a in iteranc:
                 if stoprev is not None and a < stoprev:
                     return None
                 ac = cl.read(a)  # get changeset data (we avoid object creation)
                 if path in ac[3]:  # checking the 'files' field.
                     # The file has been touched, check if the content is
                     # similar to the one we search for.
                     if fnode == mfl[ac[0]].readfast().get(path):
                         return a
             # In theory, we should never get out of that loop without a result.
             # But if manifest uses a buggy file revision (not children of the
             # one it replaces) we could. Such a buggy situation will likely
             # result is crash somewhere else at to some point.
         return lkr
 
     def isintroducedafter(self, changelogrev):
         """True if a filectx has been introduced after a given floor revision"""
         if self.linkrev() >= changelogrev:
             return True
         introrev = self._introrev(stoprev=changelogrev)
         if introrev is None:
             return False
         return introrev >= changelogrev
 
     def introrev(self):
         """return the rev of the changeset which introduced this file revision
 
         This method is different from linkrev because it take into account the
         changeset the filectx was created from. It ensures the returned
         revision is one of its ancestors. This prevents bugs from
         'linkrev-shadowing' when a file revision is used by multiple
         changesets.
         """
         return self._introrev()
 
     def _introrev(self, stoprev=None):
         """
         Same as `introrev` but, with an extra argument to limit changelog
         iteration range in some internal usecase.
 
         If `stoprev` is set, the `introrev` will not be searched past that
         `stoprev` revision and "None" might be returned. This is useful to
         limit the iteration range.
         """
         toprev = None
         attrs = vars(self)
         if '_changeid' in attrs:
             # We have a cached value already
             toprev = self._changeid
         elif '_changectx' in attrs:
             # We know which changelog entry we are coming from
             toprev = self._changectx.rev()
 
         if toprev is not None:
             return self._adjustlinkrev(toprev, inclusive=True, stoprev=stoprev)
         elif '_descendantrev' in attrs:
             introrev = self._adjustlinkrev(self._descendantrev, stoprev=stoprev)
             # be nice and cache the result of the computation
             if introrev is not None:
                 self._changeid = introrev
             return introrev
         else:
             return self.linkrev()
 
     def introfilectx(self):
         """Return filectx having identical contents, but pointing to the
         changeset revision where this filectx was introduced"""
         introrev = self.introrev()
         if self.rev() == introrev:
             return self
         return self.filectx(self.filenode(), changeid=introrev)
 
     def _parentfilectx(self, path, fileid, filelog):
         """create parent filectx keeping ancestry info for _adjustlinkrev()"""
         fctx = filectx(self._repo, path, fileid=fileid, filelog=filelog)
         if '_changeid' in vars(self) or '_changectx' in vars(self):
             # If self is associated with a changeset (probably explicitly
             # fed), ensure the created filectx is associated with a
             # changeset that is an ancestor of self.changectx.
             # This lets us later use _adjustlinkrev to get a correct link.
             fctx._descendantrev = self.rev()
             fctx._ancestrycontext = getattr(self, '_ancestrycontext', None)
         elif '_descendantrev' in vars(self):
             # Otherwise propagate _descendantrev if we have one associated.
             fctx._descendantrev = self._descendantrev
             fctx._ancestrycontext = getattr(self, '_ancestrycontext', None)
         return fctx
 
     def parents(self):
         _path = self._path
         fl = self._filelog
         parents = self._filelog.parents(self._filenode)
         pl = [(_path, node, fl) for node in parents if node != nullid]
 
         r = fl.renamed(self._filenode)
         if r:
             # - In the simple rename case, both parent are nullid, pl is empty.
             # - In case of merge, only one of the parent is null id and should
             # be replaced with the rename information. This parent is -always-
             # the first one.
             #
             # As null id have always been filtered out in the previous list
             # comprehension, inserting to 0 will always result in "replacing
             # first nullid parent with rename information.
             pl.insert(0, (r[0], r[1], self._repo.file(r[0])))
 
         return [self._parentfilectx(path, fnode, l) for path, fnode, l in pl]
 
     def p1(self):
         return self.parents()[0]
 
     def p2(self):
         p = self.parents()
         if len(p) == 2:
             return p[1]
         return filectx(self._repo, self._path, fileid=-1, filelog=self._filelog)
 
     def annotate(self, follow=False, skiprevs=None, diffopts=None):
         """Returns a list of annotateline objects for each line in the file
 
         - line.fctx is the filectx of the node where that line was last changed
         - line.lineno is the line number at the first appearance in the managed
           file
         - line.text is the data on that line (including newline character)
         """
         getlog = util.lrucachefunc(lambda x: self._repo.file(x))
 
         def parents(f):
             # Cut _descendantrev here to mitigate the penalty of lazy linkrev
             # adjustment. Otherwise, p._adjustlinkrev() would walk changelog
             # from the topmost introrev (= srcrev) down to p.linkrev() if it
             # isn't an ancestor of the srcrev.
             f._changeid
             pl = f.parents()
 
             # Don't return renamed parents if we aren't following.
             if not follow:
                 pl = [p for p in pl if p.path() == f.path()]
 
             # renamed filectx won't have a filelog yet, so set it
             # from the cache to save time
             for p in pl:
                 if not '_filelog' in p.__dict__:
                     p._filelog = getlog(p.path())
 
             return pl
 
         # use linkrev to find the first changeset where self appeared
         base = self.introfilectx()
         if getattr(base, '_ancestrycontext', None) is None:
             # it is safe to use an unfiltered repository here because we are
             # walking ancestors only.
             cl = self._repo.unfiltered().changelog
             if base.rev() is None:
                 # wctx is not inclusive, but works because _ancestrycontext
                 # is used to test filelog revisions
                 ac = cl.ancestors(
                     [p.rev() for p in base.parents()], inclusive=True
                 )
             else:
                 ac = cl.ancestors([base.rev()], inclusive=True)
             base._ancestrycontext = ac
 
         return dagop.annotate(
             base, parents, skiprevs=skiprevs, diffopts=diffopts
         )
 
     def ancestors(self, followfirst=False):
         visit = {}
         c = self
         if followfirst:
             cut = 1
         else:
             cut = None
 
         while True:
             for parent in c.parents()[:cut]:
                 visit[(parent.linkrev(), parent.filenode())] = parent
             if not visit:
                 break
             c = visit.pop(max(visit))
             yield c
 
     def decodeddata(self):
         """Returns `data()` after running repository decoding filters.
 
         This is often equivalent to how the data would be expressed on disk.
         """
         return self._repo.wwritedata(self.path(), self.data())
 
 
 class filectx(basefilectx):
     """A filecontext object makes access to data related to a particular
     filerevision convenient."""
 
     def __init__(
         self,
         repo,
         path,
         changeid=None,
         fileid=None,
         filelog=None,
         changectx=None,
     ):
         """changeid must be a revision number, if specified.
         fileid can be a file revision or node."""
         self._repo = repo
         self._path = path
 
         assert (
             changeid is not None or fileid is not None or changectx is not None
         ), b"bad args: changeid=%r, fileid=%r, changectx=%r" % (
             changeid,
             fileid,
             changectx,
         )
 
         if filelog is not None:
             self._filelog = filelog
 
         if changeid is not None:
             self._changeid = changeid
         if changectx is not None:
             self._changectx = changectx
         if fileid is not None:
             self._fileid = fileid
 
     @propertycache
     def _changectx(self):
         try:
             return self._repo[self._changeid]
         except error.FilteredRepoLookupError:
             # Linkrev may point to any revision in the repository.  When the
             # repository is filtered this may lead to `filectx` trying to build
             # `changectx` for filtered revision. In such case we fallback to
             # creating `changectx` on the unfiltered version of the reposition.
             # This fallback should not be an issue because `changectx` from
             # `filectx` are not used in complex operations that care about
             # filtering.
             #
             # This fallback is a cheap and dirty fix that prevent several
             # crashes. It does not ensure the behavior is correct. However the
             # behavior was not correct before filtering either and "incorrect
             # behavior" is seen as better as "crash"
             #
             # Linkrevs have several serious troubles with filtering that are
             # complicated to solve. Proper handling of the issue here should be
             # considered when solving linkrev issue are on the table.
             return self._repo.unfiltered()[self._changeid]
 
     def filectx(self, fileid, changeid=None):
         """opens an arbitrary revision of the file without
         opening a new filelog"""
         return filectx(
             self._repo,
             self._path,
             fileid=fileid,
             filelog=self._filelog,
             changeid=changeid,
         )
 
     def rawdata(self):
         return self._filelog.rawdata(self._filenode)
 
     def rawflags(self):
         """low-level revlog flags"""
         return self._filelog.flags(self._filerev)
 
     def data(self):
         try:
             return self._filelog.read(self._filenode)
         except error.CensoredNodeError:
             if self._repo.ui.config(b"censor", b"policy") == b"ignore":
                 return b""
             raise error.Abort(
                 _(b"censored node: %s") % short(self._filenode),
                 hint=_(b"set censor.policy to ignore errors"),
             )
 
     def size(self):
         return self._filelog.size(self._filerev)
 
     @propertycache
     def _copied(self):
         """check if file was actually renamed in this changeset revision
 
         If rename logged in file revision, we report copy for changeset only
         if file revisions linkrev points back to the changeset in question
         or both changeset parents contain different file revisions.
         """
 
         renamed = self._filelog.renamed(self._filenode)
         if not renamed:
             return None
 
         if self.rev() == self.linkrev():
             return renamed
 
         name = self.path()
         fnode = self._filenode
         for p in self._changectx.parents():
             try:
                 if fnode == p.filenode(name):
                     return None
             except error.LookupError:
                 pass
         return renamed
 
     def children(self):
         # hard for renames
         c = self._filelog.children(self._filenode)
         return [
             filectx(self._repo, self._path, fileid=x, filelog=self._filelog)
             for x in c
         ]
 
 
 class committablectx(basectx):
     """A committablectx object provides common functionality for a context that
     wants the ability to commit, e.g. workingctx or memctx."""
 
     def __init__(
         self,
         repo,
         text=b"",
         user=None,
         date=None,
         extra=None,
         changes=None,
         branch=None,
     ):
         super(committablectx, self).__init__(repo)
         self._rev = None
         self._node = None
         self._text = text
         if date:
             self._date = dateutil.parsedate(date)
         if user:
             self._user = user
         if changes:
             self._status = changes
 
         self._extra = {}
         if extra:
             self._extra = extra.copy()
         if branch is not None:
             self._extra[b'branch'] = encoding.fromlocal(branch)
         if not self._extra.get(b'branch'):
             self._extra[b'branch'] = b'default'
 
     def __bytes__(self):
         return bytes(self._parents[0]) + b"+"
 
     __str__ = encoding.strmethod(__bytes__)
 
     def __nonzero__(self):
         return True
 
     __bool__ = __nonzero__
 
     @propertycache
     def _status(self):
         return self._repo.status()
 
     @propertycache
     def _user(self):
         return self._repo.ui.username()
 
     @propertycache
     def _date(self):
         ui = self._repo.ui
         date = ui.configdate(b'devel', b'default-date')
         if date is None:
             date = dateutil.makedate()
         return date
 
     def subrev(self, subpath):
         return None
 
     def manifestnode(self):
         return None
 
     def user(self):
         return self._user or self._repo.ui.username()
 
     def date(self):
         return self._date
 
     def description(self):
         return self._text
 
     def files(self):
         return sorted(
             self._status.modified + self._status.added + self._status.removed
         )
 
     def modified(self):
         return self._status.modified
 
     def added(self):
         return self._status.added
 
     def removed(self):
         return self._status.removed
 
     def deleted(self):
         return self._status.deleted
 
     filesmodified = modified
     filesadded = added
     filesremoved = removed
 
     def branch(self):
         return encoding.tolocal(self._extra[b'branch'])
 
     def closesbranch(self):
         return b'close' in self._extra
 
     def extra(self):
         return self._extra
 
     def isinmemory(self):
         return False
 
     def tags(self):
         return []
 
     def bookmarks(self):
         b = []
         for p in self.parents():
             b.extend(p.bookmarks())
         return b
 
     def phase(self):
         phase = phases.newcommitphase(self._repo.ui)
         for p in self.parents():
             phase = max(phase, p.phase())
         return phase
 
     def hidden(self):
         return False
 
     def children(self):
         return []
 
     def flags(self, path):
         if '_manifest' in self.__dict__:
             try:
                 return self._manifest.flags(path)
             except KeyError:
                 return b''
 
         try:
             return self._flagfunc(path)
         except OSError:
             return b''
 
     def ancestor(self, c2):
         """return the "best" ancestor context of self and c2"""
         return self._parents[0].ancestor(c2)  # punt on two parents for now
 
     def ancestors(self):
         for p in self._parents:
             yield p
         for a in self._repo.changelog.ancestors(
             [p.rev() for p in self._parents]
         ):
             yield self._repo[a]
 
     def markcommitted(self, node):
         """Perform post-commit cleanup necessary after committing this ctx
 
         Specifically, this updates backing stores this working context
         wraps to reflect the fact that the changes reflected by this
         workingctx have been committed.  For example, it marks
         modified and added files as normal in the dirstate.
 
         """
 
     def dirty(self, missing=False, merge=True, branch=True):
         return False
 
 
 class workingctx(committablectx):
     """A workingctx object makes access to data related to
     the current working directory convenient.
     date - any valid date string or (unixtime, offset), or None.
     user - username string, or None.
     extra - a dictionary of extra values, or None.
     changes - a list of file lists as returned by localrepo.status()
                or None to use the repository status.
     """
 
     def __init__(
         self, repo, text=b"", user=None, date=None, extra=None, changes=None
     ):
         branch = None
         if not extra or b'branch' not in extra:
             try:
                 branch = repo.dirstate.branch()
             except UnicodeDecodeError:
                 raise error.Abort(_(b'branch name not in UTF-8!'))
         super(workingctx, self).__init__(
             repo, text, user, date, extra, changes, branch=branch
         )
 
     def __iter__(self):
         d = self._repo.dirstate
         for f in d:
             if d[f] != b'r':
                 yield f
 
     def __contains__(self, key):
         return self._repo.dirstate[key] not in b"?r"
 
     def hex(self):
         return wdirhex
 
     @propertycache
     def _parents(self):
         p = self._repo.dirstate.parents()
         if p[1] == nullid:
             p = p[:-1]
         # use unfiltered repo to delay/avoid loading obsmarkers
         unfi = self._repo.unfiltered()
         return [
             changectx(
                 self._repo, unfi.changelog.rev(n), n, maybe_filtered=False
             )
             for n in p
         ]
 
     def setparents(self, p1node, p2node=nullid):
         dirstate = self._repo.dirstate
         with dirstate.parentchange():
             copies = dirstate.setparents(p1node, p2node)
             pctx = self._repo[p1node]
             if copies:
                 # Adjust copy records, the dirstate cannot do it, it
                 # requires access to parents manifests. Preserve them
                 # only for entries added to first parent.
                 for f in copies:
                     if f not in pctx and copies[f] in pctx:
                         dirstate.copy(copies[f], f)
             if p2node == nullid:
                 for f, s in sorted(dirstate.copies().items()):
                     if f not in pctx and s not in pctx:
                         dirstate.copy(None, f)
 
     def _fileinfo(self, path):
         # populate __dict__['_manifest'] as workingctx has no _manifestdelta
         self._manifest
         return super(workingctx, self)._fileinfo(path)
 
     def _buildflagfunc(self):
         # Create a fallback function for getting file flags when the
         # filesystem doesn't support them
 
         copiesget = self._repo.dirstate.copies().get
         parents = self.parents()
         if len(parents) < 2:
             # when we have one parent, it's easy: copy from parent
             man = parents[0].manifest()
 
             def func(f):
                 f = copiesget(f, f)
                 return man.flags(f)
 
         else:
             # merges are tricky: we try to reconstruct the unstored
             # result from the merge (issue1802)
             p1, p2 = parents
             pa = p1.ancestor(p2)
             m1, m2, ma = p1.manifest(), p2.manifest(), pa.manifest()
 
             def func(f):
                 f = copiesget(f, f)  # may be wrong for merges with copies
                 fl1, fl2, fla = m1.flags(f), m2.flags(f), ma.flags(f)
                 if fl1 == fl2:
                     return fl1
                 if fl1 == fla:
                     return fl2
                 if fl2 == fla:
                     return fl1
                 return b''  # punt for conflicts
 
         return func
 
     @propertycache
     def _flagfunc(self):
         return self._repo.dirstate.flagfunc(self._buildflagfunc)
 
     def flags(self, path):
         try:
             return self._flagfunc(path)
         except OSError:
             return b''
 
     def filectx(self, path, filelog=None):
         """get a file context from the working directory"""
         return workingfilectx(
             self._repo, path, workingctx=self, filelog=filelog
         )
 
     def dirty(self, missing=False, merge=True, branch=True):
         """check whether a working directory is modified"""
         # check subrepos first
         for s in sorted(self.substate):
             if self.sub(s).dirty(missing=missing):
                 return True
         # check current working dir
         return (
             (merge and self.p2())
             or (branch and self.branch() != self.p1().branch())
             or self.modified()
             or self.added()
             or self.removed()
             or (missing and self.deleted())
         )
 
     def add(self, list, prefix=b""):
         with self._repo.wlock():
             ui, ds = self._repo.ui, self._repo.dirstate
             uipath = lambda f: ds.pathto(pathutil.join(prefix, f))
             rejected = []
             lstat = self._repo.wvfs.lstat
             for f in list:
                 # ds.pathto() returns an absolute file when this is invoked from
                 # the keyword extension.  That gets flagged as non-portable on
                 # Windows, since it contains the drive letter and colon.
                 scmutil.checkportable(ui, os.path.join(prefix, f))
                 try:
                     st = lstat(f)
                 except OSError:
                     ui.warn(_(b"%s does not exist!\n") % uipath(f))
                     rejected.append(f)
                     continue
                 limit = ui.configbytes(b'ui', b'large-file-limit')
                 if limit != 0 and st.st_size > limit:
                     ui.warn(
                         _(
                             b"%s: up to %d MB of RAM may be required "
                             b"to manage this file\n"
                             b"(use 'hg revert %s' to cancel the "
                             b"pending addition)\n"
                         )
                         % (f, 3 * st.st_size // 1000000, uipath(f))
                     )
                 if not (stat.S_ISREG(st.st_mode) or stat.S_ISLNK(st.st_mode)):
                     ui.warn(
                         _(
                             b"%s not added: only files and symlinks "
                             b"supported currently\n"
                         )
                         % uipath(f)
                     )
                     rejected.append(f)
                 elif ds[f] in b'amn':
                     ui.warn(_(b"%s already tracked!\n") % uipath(f))
                 elif ds[f] == b'r':
                     ds.normallookup(f)
                 else:
                     ds.add(f)
             return rejected
 
     def forget(self, files, prefix=b""):
         with self._repo.wlock():
             ds = self._repo.dirstate
             uipath = lambda f: ds.pathto(pathutil.join(prefix, f))
             rejected = []
             for f in files:
                 if f not in ds:
                     self._repo.ui.warn(_(b"%s not tracked!\n") % uipath(f))
                     rejected.append(f)
                 elif ds[f] != b'a':
                     ds.remove(f)
                 else:
                     ds.drop(f)
             return rejected
 
     def copy(self, source, dest):
         try:
             st = self._repo.wvfs.lstat(dest)
         except OSError as err:
             if err.errno != errno.ENOENT:
                 raise
             self._repo.ui.warn(
                 _(b"%s does not exist!\n") % self._repo.dirstate.pathto(dest)
             )
             return
         if not (stat.S_ISREG(st.st_mode) or stat.S_ISLNK(st.st_mode)):
             self._repo.ui.warn(
                 _(b"copy failed: %s is not a file or a symbolic link\n")
                 % self._repo.dirstate.pathto(dest)
             )
         else:
             with self._repo.wlock():
                 ds = self._repo.dirstate
                 if ds[dest] in b'?':
                     ds.add(dest)
                 elif ds[dest] in b'r':
                     ds.normallookup(dest)
                 ds.copy(source, dest)
 
     def match(
         self,
         pats=None,
         include=None,
         exclude=None,
         default=b'glob',
         listsubrepos=False,
         badfn=None,
         cwd=None,
     ):
         r = self._repo
         if not cwd:
             cwd = r.getcwd()
 
         # Only a case insensitive filesystem needs magic to translate user input
         # to actual case in the filesystem.
         icasefs = not util.fscasesensitive(r.root)
         return matchmod.match(
             r.root,
             cwd,
             pats,
             include,
             exclude,
             default,
             auditor=r.auditor,
             ctx=self,
             listsubrepos=listsubrepos,
             badfn=badfn,
             icasefs=icasefs,
         )
 
     def _filtersuspectsymlink(self, files):
         if not files or self._repo.dirstate._checklink:
             return files
 
         # Symlink placeholders may get non-symlink-like contents
         # via user error or dereferencing by NFS or Samba servers,
         # so we filter out any placeholders that don't look like a
         # symlink
         sane = []
         for f in files:
             if self.flags(f) == b'l':
                 d = self[f].data()
                 if (
                     d == b''
                     or len(d) >= 1024
                     or b'\n' in d
                     or stringutil.binary(d)
                 ):
                     self._repo.ui.debug(
                         b'ignoring suspect symlink placeholder "%s"\n' % f
                     )
                     continue
             sane.append(f)
         return sane
 
     def _checklookup(self, files):
         # check for any possibly clean files
         if not files:
             return [], [], []
 
         modified = []
         deleted = []
         fixup = []
         pctx = self._parents[0]
         # do a full compare of any files that might have changed
         for f in sorted(files):
             try:
                 # This will return True for a file that got replaced by a
                 # directory in the interim, but fixing that is pretty hard.
                 if (
                     f not in pctx
                     or self.flags(f) != pctx.flags(f)
                     or pctx[f].cmp(self[f])
                 ):
                     modified.append(f)
                 else:
                     fixup.append(f)
             except (IOError, OSError):
                 # A file become inaccessible in between? Mark it as deleted,
                 # matching dirstate behavior (issue5584).
                 # The dirstate has more complex behavior around whether a
                 # missing file matches a directory, etc, but we don't need to
                 # bother with that: if f has made it to this point, we're sure
                 # it's in the dirstate.
                 deleted.append(f)
 
         return modified, deleted, fixup
 
     def _poststatusfixup(self, status, fixup):
         """update dirstate for files that are actually clean"""
         poststatus = self._repo.postdsstatus()
         if fixup or poststatus:
             try:
                 oldid = self._repo.dirstate.identity()
 
                 # updating the dirstate is optional
                 # so we don't wait on the lock
                 # wlock can invalidate the dirstate, so cache normal _after_
                 # taking the lock
                 with self._repo.wlock(False):
                     if self._repo.dirstate.identity() == oldid:
                         if fixup:
                             normal = self._repo.dirstate.normal
                             for f in fixup:
                                 normal(f)
                             # write changes out explicitly, because nesting
                             # wlock at runtime may prevent 'wlock.release()'
                             # after this block from doing so for subsequent
                             # changing files
                             tr = self._repo.currenttransaction()
                             self._repo.dirstate.write(tr)
 
                         if poststatus:
                             for ps in poststatus:
                                 ps(self, status)
                     else:
                         # in this case, writing changes out breaks
                         # consistency, because .hg/dirstate was
                         # already changed simultaneously after last
                         # caching (see also issue5584 for detail)
                         self._repo.ui.debug(
                             b'skip updating dirstate: identity mismatch\n'
                         )
             except error.LockError:
                 pass
             finally:
                 # Even if the wlock couldn't be grabbed, clear out the list.
                 self._repo.clearpostdsstatus()
 
     def _dirstatestatus(self, match, ignored=False, clean=False, unknown=False):
         '''Gets the status from the dirstate -- internal use only.'''
         subrepos = []
         if b'.hgsub' in self:
             subrepos = sorted(self.substate)
         cmp, s = self._repo.dirstate.status(
             match, subrepos, ignored=ignored, clean=clean, unknown=unknown
         )
 
         # check for any possibly clean files
         fixup = []
         if cmp:
             modified2, deleted2, fixup = self._checklookup(cmp)
             s.modified.extend(modified2)
             s.deleted.extend(deleted2)
 
             if fixup and clean:
                 s.clean.extend(fixup)
 
         self._poststatusfixup(s, fixup)
 
         if match.always():
             # cache for performance
             if s.unknown or s.ignored or s.clean:
                 # "_status" is cached with list*=False in the normal route
                 self._status = scmutil.status(
                     s.modified, s.added, s.removed, s.deleted, [], [], []
                 )
             else:
                 self._status = s
 
         return s
 
     @propertycache
     def _copies(self):
         p1copies = {}
         p2copies = {}
         parents = self._repo.dirstate.parents()
         p1manifest = self._repo[parents[0]].manifest()
         p2manifest = self._repo[parents[1]].manifest()
         changedset = set(self.added()) | set(self.modified())
         narrowmatch = self._repo.narrowmatch()
         for dst, src in self._repo.dirstate.copies().items():
             if dst not in changedset or not narrowmatch(dst):
                 continue
             if src in p1manifest:
                 p1copies[dst] = src
             elif src in p2manifest:
                 p2copies[dst] = src
         return p1copies, p2copies
 
     @propertycache
     def _manifest(self):
         """generate a manifest corresponding to the values in self._status
 
         This reuse the file nodeid from parent, but we use special node
         identifiers for added and modified files. This is used by manifests
         merge to see that files are different and by update logic to avoid
         deleting newly added files.
         """
         return self._buildstatusmanifest(self._status)
 
     def _buildstatusmanifest(self, status):
         """Builds a manifest that includes the given status results."""
         parents = self.parents()
 
         man = parents[0].manifest().copy()
 
         ff = self._flagfunc
         for i, l in (
             (addednodeid, status.added),
             (modifiednodeid, status.modified),
         ):
             for f in l:
                 man[f] = i
                 try:
                     man.setflag(f, ff(f))
                 except OSError:
                     pass
 
         for f in status.deleted + status.removed:
             if f in man:
                 del man[f]
 
         return man
 
     def _buildstatus(
         self, other, s, match, listignored, listclean, listunknown
     ):
         """build a status with respect to another context
 
         This includes logic for maintaining the fast path of status when
         comparing the working directory against its parent, which is to skip
         building a new manifest if self (working directory) is not comparing
         against its parent (repo['.']).
         """
         s = self._dirstatestatus(match, listignored, listclean, listunknown)
         # Filter out symlinks that, in the case of FAT32 and NTFS filesystems,
         # might have accidentally ended up with the entire contents of the file
         # they are supposed to be linking to.
         s.modified[:] = self._filtersuspectsymlink(s.modified)
         if other != self._repo[b'.']:
             s = super(workingctx, self)._buildstatus(
                 other, s, match, listignored, listclean, listunknown
             )
         return s
 
     def _matchstatus(self, other, match):
         """override the match method with a filter for directory patterns
 
         We use inheritance to customize the match.bad method only in cases of
         workingctx since it belongs only to the working directory when
         comparing against the parent changeset.
 
         If we aren't comparing against the working directory's parent, then we
         just use the default match object sent to us.
         """
         if other != self._repo[b'.']:
 
             def bad(f, msg):
                 # 'f' may be a directory pattern from 'match.files()',
                 # so 'f not in ctx1' is not enough
                 if f not in other and not other.hasdir(f):
                     self._repo.ui.warn(
                         b'%s: %s\n' % (self._repo.dirstate.pathto(f), msg)
                     )
 
             match.bad = bad
         return match
 
     def walk(self, match):
         '''Generates matching file names.'''
         return sorted(
             self._repo.dirstate.walk(
                 self._repo.narrowmatch(match),
                 subrepos=sorted(self.substate),
                 unknown=True,
                 ignored=False,
             )
         )
 
     def matches(self, match):
         match = self._repo.narrowmatch(match)
         ds = self._repo.dirstate
         return sorted(f for f in ds.matches(match) if ds[f] != b'r')
 
     def markcommitted(self, node):
         with self._repo.dirstate.parentchange():
             for f in self.modified() + self.added():
                 self._repo.dirstate.normal(f)
             for f in self.removed():
                 self._repo.dirstate.drop(f)
             self._repo.dirstate.setparents(node)
             self._repo._quick_access_changeid_invalidate()
 
         # write changes out explicitly, because nesting wlock at
         # runtime may prevent 'wlock.release()' in 'repo.commit()'
         # from immediately doing so for subsequent changing files
         self._repo.dirstate.write(self._repo.currenttransaction())
 
         sparse.aftercommit(self._repo, node)
 
     def mergestate(self, clean=False):
         if clean:
             return mergestatemod.mergestate.clean(self._repo)
         return mergestatemod.mergestate.read(self._repo)
 
 
 class committablefilectx(basefilectx):
     """A committablefilectx provides common functionality for a file context
     that wants the ability to commit, e.g. workingfilectx or memfilectx."""
 
     def __init__(self, repo, path, filelog=None, ctx=None):
         self._repo = repo
         self._path = path
         self._changeid = None
         self._filerev = self._filenode = None
 
         if filelog is not None:
             self._filelog = filelog
         if ctx:
             self._changectx = ctx
 
     def __nonzero__(self):
         return True
 
     __bool__ = __nonzero__
 
     def linkrev(self):
         # linked to self._changectx no matter if file is modified or not
         return self.rev()
 
     def renamed(self):
         path = self.copysource()
         if not path:
             return None
         return path, self._changectx._parents[0]._manifest.get(path, nullid)
 
     def parents(self):
         '''return parent filectxs, following copies if necessary'''
 
         def filenode(ctx, path):
             return ctx._manifest.get(path, nullid)
 
         path = self._path
         fl = self._filelog
         pcl = self._changectx._parents
         renamed = self.renamed()
 
         if renamed:
             pl = [renamed + (None,)]
         else:
             pl = [(path, filenode(pcl[0], path), fl)]
 
         for pc in pcl[1:]:
             pl.append((path, filenode(pc, path), fl))
 
         return [
             self._parentfilectx(p, fileid=n, filelog=l)
             for p, n, l in pl
             if n != nullid
         ]
 
     def children(self):
         return []
 
 
 class workingfilectx(committablefilectx):
     """A workingfilectx object makes access to data related to a particular
     file in the working directory convenient."""
 
     def __init__(self, repo, path, filelog=None, workingctx=None):
         super(workingfilectx, self).__init__(repo, path, filelog, workingctx)
 
     @propertycache
     def _changectx(self):
         return workingctx(self._repo)
 
     def data(self):
         return self._repo.wread(self._path)
 
     def copysource(self):
         return self._repo.dirstate.copied(self._path)
 
     def size(self):
         return self._repo.wvfs.lstat(self._path).st_size
 
     def lstat(self):
         return self._repo.wvfs.lstat(self._path)
 
     def date(self):
         t, tz = self._changectx.date()
         try:
             return (self._repo.wvfs.lstat(self._path)[stat.ST_MTIME], tz)
         except OSError as err:
             if err.errno != errno.ENOENT:
                 raise
             return (t, tz)
 
     def exists(self):
         return self._repo.wvfs.exists(self._path)
 
     def lexists(self):
         return self._repo.wvfs.lexists(self._path)
 
     def audit(self):
         return self._repo.wvfs.audit(self._path)
 
     def cmp(self, fctx):
         """compare with other file context
 
         returns True if different than fctx.
         """
         # fctx should be a filectx (not a workingfilectx)
         # invert comparison to reuse the same code path
         return fctx.cmp(self)
 
     def remove(self, ignoremissing=False):
         """wraps unlink for a repo's working directory"""
         rmdir = self._repo.ui.configbool(b'experimental', b'removeemptydirs')
         self._repo.wvfs.unlinkpath(
             self._path, ignoremissing=ignoremissing, rmdir=rmdir
         )
 
     def write(self, data, flags, backgroundclose=False, **kwargs):
         """wraps repo.wwrite"""
         return self._repo.wwrite(
             self._path, data, flags, backgroundclose=backgroundclose, **kwargs
         )
 
     def markcopied(self, src):
         """marks this file a copy of `src`"""
         self._repo.dirstate.copy(src, self._path)
 
     def clearunknown(self):
         """Removes conflicting items in the working directory so that
         ``write()`` can be called successfully.
         """
         wvfs = self._repo.wvfs
         f = self._path
         wvfs.audit(f)
         if self._repo.ui.configbool(
             b'experimental', b'merge.checkpathconflicts'
         ):
             # remove files under the directory as they should already be
             # warned and backed up
             if wvfs.isdir(f) and not wvfs.islink(f):
                 wvfs.rmtree(f, forcibly=True)
             for p in reversed(list(pathutil.finddirs(f))):
                 if wvfs.isfileorlink(p):
                     wvfs.unlink(p)
                     break
         else:
             # don't remove files if path conflicts are not processed
             if wvfs.isdir(f) and not wvfs.islink(f):
                 wvfs.removedirs(f)
 
     def setflags(self, l, x):
         self._repo.wvfs.setflags(self._path, l, x)
 
 
 class overlayworkingctx(committablectx):
     """Wraps another mutable context with a write-back cache that can be
     converted into a commit context.
 
     self._cache[path] maps to a dict with keys: {
         'exists': bool?
         'date': date?
         'data': str?
         'flags': str?
         'copied': str? (path or None)
     }
     If `exists` is True, `flags` must be non-None and 'date' is non-None. If it
     is `False`, the file was deleted.
     """
 
     def __init__(self, repo):
         super(overlayworkingctx, self).__init__(repo)
         self.clean()
 
     def setbase(self, wrappedctx):
         self._wrappedctx = wrappedctx
         self._parents = [wrappedctx]
         # Drop old manifest cache as it is now out of date.
         # This is necessary when, e.g., rebasing several nodes with one
         # ``overlayworkingctx`` (e.g. with --collapse).
         util.clearcachedproperty(self, b'_manifest')
 
     def setparents(self, p1node, p2node=nullid):
         assert p1node == self._wrappedctx.node()
         self._parents = [self._wrappedctx, self._repo.unfiltered()[p2node]]
 
     def data(self, path):
         if self.isdirty(path):
             if self._cache[path][b'exists']:
                 if self._cache[path][b'data'] is not None:
                     return self._cache[path][b'data']
                 else:
                     # Must fallback here, too, because we only set flags.
                     return self._wrappedctx[path].data()
             else:
                 raise error.ProgrammingError(
                     b"No such file or directory: %s" % path
                 )
         else:
             return self._wrappedctx[path].data()
 
     @propertycache
     def _manifest(self):
         parents = self.parents()
         man = parents[0].manifest().copy()
 
         flag = self._flagfunc
         for path in self.added():
             man[path] = addednodeid
             man.setflag(path, flag(path))
         for path in self.modified():
             man[path] = modifiednodeid
             man.setflag(path, flag(path))
         for path in self.removed():
             del man[path]
         return man
 
     @propertycache
     def _flagfunc(self):
         def f(path):
             return self._cache[path][b'flags']
 
         return f
 
     def files(self):
         return sorted(self.added() + self.modified() + self.removed())
 
     def modified(self):
         return [
             f
             for f in self._cache.keys()
             if self._cache[f][b'exists'] and self._existsinparent(f)
         ]
 
     def added(self):
         return [
             f
             for f in self._cache.keys()
             if self._cache[f][b'exists'] and not self._existsinparent(f)
         ]
 
     def removed(self):
         return [
             f
             for f in self._cache.keys()
             if not self._cache[f][b'exists'] and self._existsinparent(f)
         ]
 
     def p1copies(self):
         copies = {}
         narrowmatch = self._repo.narrowmatch()
         for f in self._cache.keys():
             if not narrowmatch(f):
                 continue
             copies.pop(f, None)  # delete if it exists
             source = self._cache[f][b'copied']
             if source:
                 copies[f] = source
         return copies
 
     def p2copies(self):
         copies = {}
         narrowmatch = self._repo.narrowmatch()
         for f in self._cache.keys():
             if not narrowmatch(f):
                 continue
             copies.pop(f, None)  # delete if it exists
             source = self._cache[f][b'copied']
             if source:
                 copies[f] = source
         return copies
 
     def isinmemory(self):
         return True
 
     def filedate(self, path):
         if self.isdirty(path):
             return self._cache[path][b'date']
         else:
             return self._wrappedctx[path].date()
 
     def markcopied(self, path, origin):
         self._markdirty(
             path,
             exists=True,
             date=self.filedate(path),
             flags=self.flags(path),
             copied=origin,
         )
 
     def copydata(self, path):
         if self.isdirty(path):
             return self._cache[path][b'copied']
         else:
             return None
 
     def flags(self, path):
         if self.isdirty(path):
             if self._cache[path][b'exists']:
                 return self._cache[path][b'flags']
             else:
                 raise error.ProgrammingError(
                     b"No such file or directory: %s" % path
                 )
         else:
             return self._wrappedctx[path].flags()
 
     def __contains__(self, key):
         if key in self._cache:
             return self._cache[key][b'exists']
         return key in self.p1()
 
     def _existsinparent(self, path):
         try:
             # ``commitctx` raises a ``ManifestLookupError`` if a path does not
             # exist, unlike ``workingctx``, which returns a ``workingfilectx``
             # with an ``exists()`` function.
             self._wrappedctx[path]
             return True
         except error.ManifestLookupError:
             return False
 
     def _auditconflicts(self, path):
         """Replicates conflict checks done by wvfs.write().
 
         Since we never write to the filesystem and never call `applyupdates` in
         IMM, we'll never check that a path is actually writable -- e.g., because
         it adds `a/foo`, but `a` is actually a file in the other commit.
         """
 
         def fail(path, component):
             # p1() is the base and we're receiving "writes" for p2()'s
             # files.
             if b'l' in self.p1()[component].flags():
                 raise error.Abort(
                     b"error: %s conflicts with symlink %s "
                     b"in %d." % (path, component, self.p1().rev())
                 )
             else:
                 raise error.Abort(
                     b"error: '%s' conflicts with file '%s' in "
                     b"%d." % (path, component, self.p1().rev())
                 )
 
         # Test that each new directory to be created to write this path from p2
         # is not a file in p1.
         components = path.split(b'/')
         for i in pycompat.xrange(len(components)):
             component = b"/".join(components[0:i])
             if component in self:
                 fail(path, component)
 
         # Test the other direction -- that this path from p2 isn't a directory
         # in p1 (test that p1 doesn't have any paths matching `path/*`).
         match = self.match([path], default=b'path')
         mfiles = list(self.p1().manifest().walk(match))
         if len(mfiles) > 0:
             if len(mfiles) == 1 and mfiles[0] == path:
                 return
             # omit the files which are deleted in current IMM wctx
             mfiles = [m for m in mfiles if m in self]
             if not mfiles:
                 return
             raise error.Abort(
                 b"error: file '%s' cannot be written because "
                 b" '%s/' is a directory in %s (containing %d "
                 b"entries: %s)"
                 % (path, path, self.p1(), len(mfiles), b', '.join(mfiles))
             )
 
     def write(self, path, data, flags=b'', **kwargs):
         if data is None:
             raise error.ProgrammingError(b"data must be non-None")
         self._auditconflicts(path)
         self._markdirty(
             path, exists=True, data=data, date=dateutil.makedate(), flags=flags
         )
 
     def setflags(self, path, l, x):
         flag = b''
         if l:
             flag = b'l'
         elif x:
             flag = b'x'
         self._markdirty(path, exists=True, date=dateutil.makedate(), flags=flag)
 
     def remove(self, path):
         self._markdirty(path, exists=False)
 
     def exists(self, path):
         """exists behaves like `lexists`, but needs to follow symlinks and
         return False if they are broken.
         """
         if self.isdirty(path):
             # If this path exists and is a symlink, "follow" it by calling
             # exists on the destination path.
             if (
                 self._cache[path][b'exists']
                 and b'l' in self._cache[path][b'flags']
             ):
                 return self.exists(self._cache[path][b'data'].strip())
             else:
                 return self._cache[path][b'exists']
 
         return self._existsinparent(path)
 
     def lexists(self, path):
         """lexists returns True if the path exists"""
         if self.isdirty(path):
             return self._cache[path][b'exists']
 
         return self._existsinparent(path)
 
     def size(self, path):
         if self.isdirty(path):
             if self._cache[path][b'exists']:
                 return len(self._cache[path][b'data'])
             else:
                 raise error.ProgrammingError(
                     b"No such file or directory: %s" % path
                 )
         return self._wrappedctx[path].size()
 
     def tomemctx(
         self,
         text,
         branch=None,
         extra=None,
         date=None,
         parents=None,
         user=None,
         editor=None,
     ):
         """Converts this ``overlayworkingctx`` into a ``memctx`` ready to be
         committed.
 
         ``text`` is the commit message.
         ``parents`` (optional) are rev numbers.
         """
         # Default parents to the wrapped context if not passed.
         if parents is None:
             parents = self.parents()
             if len(parents) == 1:
                 parents = (parents[0], None)
 
         # ``parents`` is passed as rev numbers; convert to ``commitctxs``.
         if parents[1] is None:
             parents = (self._repo[parents[0]], None)
         else:
             parents = (self._repo[parents[0]], self._repo[parents[1]])
 
         files = self.files()
 
         def getfile(repo, memctx, path):
             if self._cache[path][b'exists']:
                 return memfilectx(
                     repo,
                     memctx,
                     path,
                     self._cache[path][b'data'],
                     b'l' in self._cache[path][b'flags'],
                     b'x' in self._cache[path][b'flags'],
                     self._cache[path][b'copied'],
                 )
             else:
                 # Returning None, but including the path in `files`, is
                 # necessary for memctx to register a deletion.
                 return None
 
         if branch is None:
             branch = self._wrappedctx.branch()
 
         return memctx(
             self._repo,
             parents,
             text,
             files,
             getfile,
             date=date,
             extra=extra,
             user=user,
             branch=branch,
             editor=editor,
         )
 
     def tomemctx_for_amend(self, precursor):
         extra = precursor.extra().copy()
         extra[b'amend_source'] = precursor.hex()
         return self.tomemctx(
             text=precursor.description(),
             branch=precursor.branch(),
             extra=extra,
             date=precursor.date(),
             user=precursor.user(),
         )
 
     def isdirty(self, path):
         return path in self._cache
 
     def clean(self):
         self._mergestate = None
         self._cache = {}
 
     def _compact(self):
         """Removes keys from the cache that are actually clean, by comparing
         them with the underlying context.
 
         This can occur during the merge process, e.g. by passing --tool :local
         to resolve a conflict.
         """
         keys = []
         # This won't be perfect, but can help performance significantly when
         # using things like remotefilelog.
         scmutil.prefetchfiles(
             self.repo(),
             [
                 (
                     self.p1().rev(),
                     scmutil.matchfiles(self.repo(), self._cache.keys()),
                 )
             ],
         )
 
         for path in self._cache.keys():
             cache = self._cache[path]
             try:
                 underlying = self._wrappedctx[path]
                 if (
                     underlying.data() == cache[b'data']
                     and underlying.flags() == cache[b'flags']
                 ):
                     keys.append(path)
             except error.ManifestLookupError:
                 # Path not in the underlying manifest (created).
                 continue
 
         for path in keys:
             del self._cache[path]
         return keys
 
     def _markdirty(
         self, path, exists, data=None, date=None, flags=b'', copied=None
     ):
         # data not provided, let's see if we already have some; if not, let's
         # grab it from our underlying context, so that we always have data if
         # the file is marked as existing.
         if exists and data is None:
             oldentry = self._cache.get(path) or {}
             data = oldentry.get(b'data')
             if data is None:
                 data = self._wrappedctx[path].data()
 
         self._cache[path] = {
             b'exists': exists,
             b'data': data,
             b'date': date,
             b'flags': flags,
             b'copied': copied,
         }
         util.clearcachedproperty(self, b'_manifest')
 
     def filectx(self, path, filelog=None):
         return overlayworkingfilectx(
             self._repo, path, parent=self, filelog=filelog
         )
 
     def mergestate(self, clean=False):
         if clean or self._mergestate is None:
             self._mergestate = mergestatemod.memmergestate(self._repo)
         return self._mergestate
 
 
 class overlayworkingfilectx(committablefilectx):
     """Wrap a ``workingfilectx`` but intercepts all writes into an in-memory
     cache, which can be flushed through later by calling ``flush()``."""
 
     def __init__(self, repo, path, filelog=None, parent=None):
         super(overlayworkingfilectx, self).__init__(repo, path, filelog, parent)
         self._repo = repo
         self._parent = parent
         self._path = path
 
     def cmp(self, fctx):
         return self.data() != fctx.data()
 
     def changectx(self):
         return self._parent
 
     def data(self):
         return self._parent.data(self._path)
 
     def date(self):
         return self._parent.filedate(self._path)
 
     def exists(self):
         return self.lexists()
 
     def lexists(self):
         return self._parent.exists(self._path)
 
     def copysource(self):
         return self._parent.copydata(self._path)
 
     def size(self):
         return self._parent.size(self._path)
 
     def markcopied(self, origin):
         self._parent.markcopied(self._path, origin)
 
     def audit(self):
         pass
 
     def flags(self):
         return self._parent.flags(self._path)
 
     def setflags(self, islink, isexec):
         return self._parent.setflags(self._path, islink, isexec)
 
     def write(self, data, flags, backgroundclose=False, **kwargs):
         return self._parent.write(self._path, data, flags, **kwargs)
 
     def remove(self, ignoremissing=False):
         return self._parent.remove(self._path)
 
     def clearunknown(self):
         pass
 
 
 class workingcommitctx(workingctx):
     """A workingcommitctx object makes access to data related to
     the revision being committed convenient.
 
     This hides changes in the working directory, if they aren't
     committed in this context.
     """
 
     def __init__(
         self, repo, changes, text=b"", user=None, date=None, extra=None
     ):
         super(workingcommitctx, self).__init__(
             repo, text, user, date, extra, changes
         )
 
     def _dirstatestatus(self, match, ignored=False, clean=False, unknown=False):
         """Return matched files only in ``self._status``
 
         Uncommitted files appear "clean" via this context, even if
         they aren't actually so in the working directory.
         """
         if clean:
             clean = [f for f in self._manifest if f not in self._changedset]
         else:
             clean = []
         return scmutil.status(
             [f for f in self._status.modified if match(f)],
             [f for f in self._status.added if match(f)],
             [f for f in self._status.removed if match(f)],
             [],
             [],
             [],
             clean,
         )
 
     @propertycache
     def _changedset(self):
         """Return the set of files changed in this context"""
         changed = set(self._status.modified)
         changed.update(self._status.added)
         changed.update(self._status.removed)
         return changed
 
 
 def makecachingfilectxfn(func):
     """Create a filectxfn that caches based on the path.
 
     We can't use util.cachefunc because it uses all arguments as the cache
     key and this creates a cycle since the arguments include the repo and
     memctx.
     """
     cache = {}
 
     def getfilectx(repo, memctx, path):
         if path not in cache:
             cache[path] = func(repo, memctx, path)
         return cache[path]
 
     return getfilectx
 
 
 def memfilefromctx(ctx):
     """Given a context return a memfilectx for ctx[path]
 
     This is a convenience method for building a memctx based on another
     context.
     """
 
     def getfilectx(repo, memctx, path):
         fctx = ctx[path]
         copysource = fctx.copysource()
         return memfilectx(
             repo,
             memctx,
             path,
             fctx.data(),
             islink=fctx.islink(),
             isexec=fctx.isexec(),
             copysource=copysource,
         )
 
     return getfilectx
 
 
 def memfilefrompatch(patchstore):
     """Given a patch (e.g. patchstore object) return a memfilectx
 
     This is a convenience method for building a memctx based on a patchstore.
     """
 
     def getfilectx(repo, memctx, path):
         data, mode, copysource = patchstore.getfile(path)
         if data is None:
             return None
         islink, isexec = mode
         return memfilectx(
             repo,
             memctx,
             path,
             data,
             islink=islink,
             isexec=isexec,
             copysource=copysource,
         )
 
     return getfilectx
 
 
 class memctx(committablectx):
     """Use memctx to perform in-memory commits via localrepo.commitctx().
 
     Revision information is supplied at initialization time while
     related files data and is made available through a callback
     mechanism.  'repo' is the current localrepo, 'parents' is a
     sequence of two parent revisions identifiers (pass None for every
     missing parent), 'text' is the commit message and 'files' lists
     names of files touched by the revision (normalized and relative to
     repository root).
 
     filectxfn(repo, memctx, path) is a callable receiving the
     repository, the current memctx object and the normalized path of
     requested file, relative to repository root. It is fired by the
     commit function for every file in 'files', but calls order is
     undefined. If the file is available in the revision being
     committed (updated or added), filectxfn returns a memfilectx
     object. If the file was removed, filectxfn return None for recent
     Mercurial. Moved files are represented by marking the source file
     removed and the new file added with copy information (see
     memfilectx).
 
     user receives the committer name and defaults to current
     repository username, date is the commit date in any format
     supported by dateutil.parsedate() and defaults to current date, extra
     is a dictionary of metadata or is left empty.
     """
 
     # Mercurial <= 3.1 expects the filectxfn to raise IOError for missing files.
     # Extensions that need to retain compatibility across Mercurial 3.1 can use
     # this field to determine what to do in filectxfn.
     _returnnoneformissingfiles = True
 
     def __init__(
         self,
         repo,
         parents,
         text,
         files,
         filectxfn,
         user=None,
         date=None,
         extra=None,
         branch=None,
         editor=None,
     ):
         super(memctx, self).__init__(
             repo, text, user, date, extra, branch=branch
         )
         self._rev = None
         self._node = None
         parents = [(p or nullid) for p in parents]
         p1, p2 = parents
         self._parents = [self._repo[p] for p in (p1, p2)]
         files = sorted(set(files))
         self._files = files
         self.substate = {}
 
         if isinstance(filectxfn, patch.filestore):
             filectxfn = memfilefrompatch(filectxfn)
         elif not callable(filectxfn):
             # if store is not callable, wrap it in a function
             filectxfn = memfilefromctx(filectxfn)
 
         # memoizing increases performance for e.g. vcs convert scenarios.
         self._filectxfn = makecachingfilectxfn(filectxfn)
 
         if editor:
             self._text = editor(self._repo, self, [])
             self._repo.savecommitmessage(self._text)
 
     def filectx(self, path, filelog=None):
         """get a file context from the working directory
 
         Returns None if file doesn't exist and should be removed."""
         return self._filectxfn(self._repo, self, path)
 
     def commit(self):
         """commit context to the repo"""
         return self._repo.commitctx(self)
 
     @propertycache
     def _manifest(self):
         """generate a manifest based on the return values of filectxfn"""
 
         # keep this simple for now; just worry about p1
         pctx = self._parents[0]
         man = pctx.manifest().copy()
 
         for f in self._status.modified:
             man[f] = modifiednodeid
 
         for f in self._status.added:
             man[f] = addednodeid
 
         for f in self._status.removed:
             if f in man:
                 del man[f]
 
         return man
 
     @propertycache
     def _status(self):
         """Calculate exact status from ``files`` specified at construction"""
         man1 = self.p1().manifest()
         p2 = self._parents[1]
         # "1 < len(self._parents)" can't be used for checking
         # existence of the 2nd parent, because "memctx._parents" is
         # explicitly initialized by the list, of which length is 2.
-        if p2.node() != nullid:
+        if p2.rev() != nullrev:
             man2 = p2.manifest()
             managing = lambda f: f in man1 or f in man2
         else:
             managing = lambda f: f in man1
 
         modified, added, removed = [], [], []
         for f in self._files:
             if not managing(f):
                 added.append(f)
             elif self[f]:
                 modified.append(f)
             else:
                 removed.append(f)
 
         return scmutil.status(modified, added, removed, [], [], [], [])
 
     def parents(self):
-        if self._parents[1].node() == nullid:
+        if self._parents[1].rev() == nullrev:
             return [self._parents[0]]
         return self._parents
 
 
 class memfilectx(committablefilectx):
     """memfilectx represents an in-memory file to commit.
 
     See memctx and committablefilectx for more details.
     """
 
     def __init__(
         self,
         repo,
         changectx,
         path,
         data,
         islink=False,
         isexec=False,
         copysource=None,
     ):
         """
         path is the normalized file path relative to repository root.
         data is the file content as a string.
         islink is True if the file is a symbolic link.
         isexec is True if the file is executable.
         copied is the source file path if current file was copied in the
         revision being committed, or None."""
         super(memfilectx, self).__init__(repo, path, None, changectx)
         self._data = data
         if islink:
             self._flags = b'l'
         elif isexec:
             self._flags = b'x'
         else:
             self._flags = b''
         self._copysource = copysource
 
     def copysource(self):
         return self._copysource
 
     def cmp(self, fctx):
         return self.data() != fctx.data()
 
     def data(self):
         return self._data
 
     def remove(self, ignoremissing=False):
         """wraps unlink for a repo's working directory"""
         # need to figure out what to do here
         del self._changectx[self._path]
 
     def write(self, data, flags, **kwargs):
         """wraps repo.wwrite"""
         self._data = data
 
 
 class metadataonlyctx(committablectx):
     """Like memctx but it's reusing the manifest of different commit.
     Intended to be used by lightweight operations that are creating
     metadata-only changes.
 
     Revision information is supplied at initialization time.  'repo' is the
     current localrepo, 'ctx' is original revision which manifest we're reuisng
     'parents' is a sequence of two parent revisions identifiers (pass None for
     every missing parent), 'text' is the commit.
 
     user receives the committer name and defaults to current repository
     username, date is the commit date in any format supported by
     dateutil.parsedate() and defaults to current date, extra is a dictionary of
     metadata or is left empty.
     """
 
     def __init__(
         self,
         repo,
         originalctx,
         parents=None,
         text=None,
         user=None,
         date=None,
         extra=None,
         editor=None,
     ):
         if text is None:
             text = originalctx.description()
         super(metadataonlyctx, self).__init__(repo, text, user, date, extra)
         self._rev = None
         self._node = None
         self._originalctx = originalctx
         self._manifestnode = originalctx.manifestnode()
         if parents is None:
             parents = originalctx.parents()
         else:
             parents = [repo[p] for p in parents if p is not None]
         parents = parents[:]
         while len(parents) < 2:
             parents.append(repo[nullrev])
         p1, p2 = self._parents = parents
 
         # sanity check to ensure that the reused manifest parents are
         # manifests of our commit parents
         mp1, mp2 = self.manifestctx().parents
         if p1 != nullid and p1.manifestnode() != mp1:
             raise RuntimeError(
                 r"can't reuse the manifest: its p1 "
                 r"doesn't match the new ctx p1"
             )
         if p2 != nullid and p2.manifestnode() != mp2:
             raise RuntimeError(
                 r"can't reuse the manifest: "
                 r"its p2 doesn't match the new ctx p2"
             )
 
         self._files = originalctx.files()
         self.substate = {}
 
         if editor:
             self._text = editor(self._repo, self, [])
             self._repo.savecommitmessage(self._text)
 
     def manifestnode(self):
         return self._manifestnode
 
     @property
     def _manifestctx(self):
         return self._repo.manifestlog[self._manifestnode]
 
     def filectx(self, path, filelog=None):
         return self._originalctx.filectx(path, filelog=filelog)
 
     def commit(self):
         """commit context to the repo"""
         return self._repo.commitctx(self)
 
     @property
     def _manifest(self):
         return self._originalctx.manifest()
 
     @propertycache
     def _status(self):
         """Calculate exact status from ``files`` specified in the ``origctx``
         and parents manifests.
         """
         man1 = self.p1().manifest()
         p2 = self._parents[1]
         # "1 < len(self._parents)" can't be used for checking
         # existence of the 2nd parent, because "metadataonlyctx._parents" is
         # explicitly initialized by the list, of which length is 2.
-        if p2.node() != nullid:
+        if p2.rev() != nullrev:
             man2 = p2.manifest()
             managing = lambda f: f in man1 or f in man2
         else:
             managing = lambda f: f in man1
 
         modified, added, removed = [], [], []
         for f in self._files:
             if not managing(f):
                 added.append(f)
             elif f in self:
                 modified.append(f)
             else:
                 removed.append(f)
 
         return scmutil.status(modified, added, removed, [], [], [], [])
 
 
 class arbitraryfilectx(object):
     """Allows you to use filectx-like functions on a file in an arbitrary
     location on disk, possibly not in the working directory.
     """
 
     def __init__(self, path, repo=None):
         # Repo is optional because contrib/simplemerge uses this class.
         self._repo = repo
         self._path = path
 
     def cmp(self, fctx):
         # filecmp follows symlinks whereas `cmp` should not, so skip the fast
         # path if either side is a symlink.
         symlinks = b'l' in self.flags() or b'l' in fctx.flags()
         if not symlinks and isinstance(fctx, workingfilectx) and self._repo:
             # Add a fast-path for merge if both sides are disk-backed.
             # Note that filecmp uses the opposite return values (True if same)
             # from our cmp functions (True if different).
             return not filecmp.cmp(self.path(), self._repo.wjoin(fctx.path()))
         return self.data() != fctx.data()
 
     def path(self):
         return self._path
 
     def flags(self):
         return b''
 
     def data(self):
         return util.readfile(self._path)
 
     def decodeddata(self):
         with open(self._path, b"rb") as f:
             return f.read()
 
     def remove(self):
         util.unlink(self._path)
 
     def write(self, data, flags, **kwargs):
         assert not flags
         with open(self._path, b"wb") as f:
             f.write(data)

mercurial/copies.py

@@ -1,1308 +1,1308 b''
 # coding: utf8
 # copies.py - copy detection for Mercurial
 #
 # Copyright 2008 Olivia Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import collections
 import os
 
 from .i18n import _
 from .node import (
     nullid,
     nullrev,
 )
 
 from . import (
     match as matchmod,
     pathutil,
     policy,
     pycompat,
     util,
 )
 
 
 from .utils import stringutil
 
 from .revlogutils import (
     flagutil,
     sidedata as sidedatamod,
 )
 
 rustmod = policy.importrust("copy_tracing")
 
 
 def _filter(src, dst, t):
     """filters out invalid copies after chaining"""
 
     # When _chain()'ing copies in 'a' (from 'src' via some other commit 'mid')
     # with copies in 'b' (from 'mid' to 'dst'), we can get the different cases
     # in the following table (not including trivial cases). For example, case 6
     # is where a file existed in 'src' and remained under that name in 'mid' and
     # then was renamed between 'mid' and 'dst'.
     #
     # case src mid dst result
     #   1   x   y   -    -
     #   2   x   y   y   x->y
     #   3   x   y   x    -
     #   4   x   y   z   x->z
     #   5   -   x   y    -
     #   6   x   x   y   x->y
     #
     # _chain() takes care of chaining the copies in 'a' and 'b', but it
     # cannot tell the difference between cases 1 and 2, between 3 and 4, or
     # between 5 and 6, so it includes all cases in its result.
     # Cases 1, 3, and 5 are then removed by _filter().
 
     for k, v in list(t.items()):
         if k == v:  # case 3
             del t[k]
         elif v not in src:  # case 5
             # remove copies from files that didn't exist
             del t[k]
         elif k not in dst:  # case 1
             # remove copies to files that were then removed
             del t[k]
 
 
 def _chain(prefix, suffix):
     """chain two sets of copies 'prefix' and 'suffix'"""
     result = prefix.copy()
     for key, value in pycompat.iteritems(suffix):
         result[key] = prefix.get(value, value)
     return result
 
 
 def _tracefile(fctx, am, basemf):
     """return file context that is the ancestor of fctx present in ancestor
     manifest am
 
     Note: we used to try and stop after a given limit, however checking if that
     limit is reached turned out to be very expensive. we are better off
     disabling that feature."""
 
     for f in fctx.ancestors():
         path = f.path()
         if am.get(path, None) == f.filenode():
             return path
         if basemf and basemf.get(path, None) == f.filenode():
             return path
 
 
 def _dirstatecopies(repo, match=None):
     ds = repo.dirstate
     c = ds.copies().copy()
     for k in list(c):
         if ds[k] not in b'anm' or (match and not match(k)):
             del c[k]
     return c
 
 
 def _computeforwardmissing(a, b, match=None):
     """Computes which files are in b but not a.
     This is its own function so extensions can easily wrap this call to see what
     files _forwardcopies is about to process.
     """
     ma = a.manifest()
     mb = b.manifest()
     return mb.filesnotin(ma, match=match)
 
 
 def usechangesetcentricalgo(repo):
     """Checks if we should use changeset-centric copy algorithms"""
     if repo.filecopiesmode == b'changeset-sidedata':
         return True
     readfrom = repo.ui.config(b'experimental', b'copies.read-from')
     changesetsource = (b'changeset-only', b'compatibility')
     return readfrom in changesetsource
 
 
 def _committedforwardcopies(a, b, base, match):
     """Like _forwardcopies(), but b.rev() cannot be None (working copy)"""
     # files might have to be traced back to the fctx parent of the last
     # one-side-only changeset, but not further back than that
     repo = a._repo
 
     if usechangesetcentricalgo(repo):
         return _changesetforwardcopies(a, b, match)
 
     debug = repo.ui.debugflag and repo.ui.configbool(b'devel', b'debug.copies')
     dbg = repo.ui.debug
     if debug:
         dbg(b'debug.copies:    looking into rename from %s to %s\n' % (a, b))
     am = a.manifest()
     basemf = None if base is None else base.manifest()
 
     # find where new files came from
     # we currently don't try to find where old files went, too expensive
     # this means we can miss a case like 'hg rm b; hg cp a b'
     cm = {}
 
     # Computing the forward missing is quite expensive on large manifests, since
     # it compares the entire manifests. We can optimize it in the common use
     # case of computing what copies are in a commit versus its parent (like
     # during a rebase or histedit). Note, we exclude merge commits from this
     # optimization, since the ctx.files() for a merge commit is not correct for
     # this comparison.
     forwardmissingmatch = match
-    if b.p1() == a and b.p2().node() == nullid:
+    if b.p1() == a and b.p2().rev() == nullrev:
         filesmatcher = matchmod.exact(b.files())
         forwardmissingmatch = matchmod.intersectmatchers(match, filesmatcher)
     if repo.ui.configbool(b'devel', b'copy-tracing.trace-all-files'):
         missing = list(b.walk(match))
         # _computeforwardmissing(a, b, match=forwardmissingmatch)
         if debug:
             dbg(b'debug.copies:      searching all files: %d\n' % len(missing))
     else:
         missing = _computeforwardmissing(a, b, match=forwardmissingmatch)
         if debug:
             dbg(
                 b'debug.copies:      missing files to search: %d\n'
                 % len(missing)
             )
 
     ancestrycontext = a._repo.changelog.ancestors([b.rev()], inclusive=True)
 
     for f in sorted(missing):
         if debug:
             dbg(b'debug.copies:        tracing file: %s\n' % f)
         fctx = b[f]
         fctx._ancestrycontext = ancestrycontext
 
         if debug:
             start = util.timer()
         opath = _tracefile(fctx, am, basemf)
         if opath:
             if debug:
                 dbg(b'debug.copies:          rename of: %s\n' % opath)
             cm[f] = opath
         if debug:
             dbg(
                 b'debug.copies:          time: %f seconds\n'
                 % (util.timer() - start)
             )
     return cm
 
 
 def _revinfo_getter(repo, match):
     """returns a function that returns the following data given a <rev>"
 
     * p1: revision number of first parent
     * p2: revision number of first parent
     * changes: a ChangingFiles object
     """
     cl = repo.changelog
     parents = cl.parentrevs
     flags = cl.flags
 
     HASCOPIESINFO = flagutil.REVIDX_HASCOPIESINFO
 
     changelogrevision = cl.changelogrevision
 
     if rustmod is not None:
 
         def revinfo(rev):
             p1, p2 = parents(rev)
             if flags(rev) & HASCOPIESINFO:
                 raw = changelogrevision(rev)._sidedata.get(sidedatamod.SD_FILES)
             else:
                 raw = None
             return (p1, p2, raw)
 
     else:
 
         def revinfo(rev):
             p1, p2 = parents(rev)
             if flags(rev) & HASCOPIESINFO:
                 changes = changelogrevision(rev).changes
             else:
                 changes = None
             return (p1, p2, changes)
 
     return revinfo
 
 
 def cached_is_ancestor(is_ancestor):
     """return a cached version of is_ancestor"""
     cache = {}
 
     def _is_ancestor(anc, desc):
         if anc > desc:
             return False
         elif anc == desc:
             return True
         key = (anc, desc)
         ret = cache.get(key)
         if ret is None:
             ret = cache[key] = is_ancestor(anc, desc)
         return ret
 
     return _is_ancestor
 
 
 def _changesetforwardcopies(a, b, match):
     if a.rev() in (nullrev, b.rev()):
         return {}
 
     repo = a.repo().unfiltered()
     children = {}
 
     cl = repo.changelog
     isancestor = cl.isancestorrev
 
     # To track rename from "A" to B, we need to gather all parent → children
     # edges that are contains in `::B` but not in `::A`.
     #
     #
     # To do so, we need to gather all revisions exclusive¹ to "B" (ie¹: `::b -
     # ::a`) and also all the "roots point", ie the parents of the exclusive set
     # that belong to ::a. These are exactly all the revisions needed to express
     # the parent → children we need to combine.
     #
     # [1] actually, we need to gather all the edges within `(::a)::b`, ie:
     # excluding paths that leads to roots that are not ancestors of `a`. We
     # keep this out of the explanation because it is hard enough without this special case..
 
     parents = cl._uncheckedparentrevs
     graph_roots = (nullrev, nullrev)
 
     ancestors = cl.ancestors([a.rev()], inclusive=True)
     revs = cl.findmissingrevs(common=[a.rev()], heads=[b.rev()])
     roots = set()
     has_graph_roots = False
     multi_thread = repo.ui.configbool(b'devel', b'copy-tracing.multi-thread')
 
     # iterate over `only(B, A)`
     for r in revs:
         ps = parents(r)
         if ps == graph_roots:
             has_graph_roots = True
         else:
             p1, p2 = ps
 
             # find all the "root points" (see larger comment above)
             if p1 != nullrev and p1 in ancestors:
                 roots.add(p1)
             if p2 != nullrev and p2 in ancestors:
                 roots.add(p2)
     if not roots:
         # no common revision to track copies from
         return {}
     if has_graph_roots:
         # this deal with the special case mentionned in the [1] footnotes. We
         # must filter out revisions that leads to non-common graphroots.
         roots = list(roots)
         m = min(roots)
         h = [b.rev()]
         roots_to_head = cl.reachableroots(m, h, roots, includepath=True)
         roots_to_head = set(roots_to_head)
         revs = [r for r in revs if r in roots_to_head]
 
     if repo.filecopiesmode == b'changeset-sidedata':
         # When using side-data, we will process the edges "from" the children.
         # We iterate over the childre, gathering previous collected data for
         # the parents. Do know when the parents data is no longer necessary, we
         # keep a counter of how many children each revision has.
         #
         # An interresting property of `children_count` is that it only contains
         # revision that will be relevant for a edge of the graph. So if a
         # children has parent not in `children_count`, that edges should not be
         # processed.
         children_count = dict((r, 0) for r in roots)
         for r in revs:
             for p in cl.parentrevs(r):
                 if p == nullrev:
                     continue
                 children_count[r] = 0
                 if p in children_count:
                     children_count[p] += 1
         revinfo = _revinfo_getter(repo, match)
         return _combine_changeset_copies(
             revs,
             children_count,
             b.rev(),
             revinfo,
             match,
             isancestor,
             multi_thread,
         )
     else:
         # When not using side-data, we will process the edges "from" the parent.
         # so we need a full mapping of the parent -> children relation.
         children = dict((r, []) for r in roots)
         for r in revs:
             for p in cl.parentrevs(r):
                 if p == nullrev:
                     continue
                 children[r] = []
                 if p in children:
                     children[p].append(r)
         x = revs.pop()
         assert x == b.rev()
         revs.extend(roots)
         revs.sort()
 
         revinfo = _revinfo_getter_extra(repo)
         return _combine_changeset_copies_extra(
             revs, children, b.rev(), revinfo, match, isancestor
         )
 
 
 def _combine_changeset_copies(
     revs, children_count, targetrev, revinfo, match, isancestor, multi_thread
 ):
     """combine the copies information for each item of iterrevs
 
     revs: sorted iterable of revision to visit
     children_count: a {parent: <number-of-relevant-children>} mapping.
     targetrev: the final copies destination revision (not in iterrevs)
     revinfo(rev): a function that return (p1, p2, p1copies, p2copies, removed)
     match: a matcher
 
     It returns the aggregated copies information for `targetrev`.
     """
 
     alwaysmatch = match.always()
 
     if rustmod is not None:
         final_copies = rustmod.combine_changeset_copies(
             list(revs), children_count, targetrev, revinfo, multi_thread
         )
     else:
         isancestor = cached_is_ancestor(isancestor)
 
         all_copies = {}
         # iterate over all the "children" side of copy tracing "edge"
         for current_rev in revs:
             p1, p2, changes = revinfo(current_rev)
             current_copies = None
             # iterate over all parents to chain the existing data with the
             # data from the parent → child edge.
             for parent, parent_rev in ((1, p1), (2, p2)):
                 if parent_rev == nullrev:
                     continue
                 remaining_children = children_count.get(parent_rev)
                 if remaining_children is None:
                     continue
                 remaining_children -= 1
                 children_count[parent_rev] = remaining_children
                 if remaining_children:
                     copies = all_copies.get(parent_rev, None)
                 else:
                     copies = all_copies.pop(parent_rev, None)
 
                 if copies is None:
                     # this is a root
                     newcopies = copies = {}
                 elif remaining_children:
                     newcopies = copies.copy()
                 else:
                     newcopies = copies
                 # chain the data in the edge with the existing data
                 if changes is not None:
                     childcopies = {}
                     if parent == 1:
                         childcopies = changes.copied_from_p1
                     elif parent == 2:
                         childcopies = changes.copied_from_p2
 
                     if childcopies:
                         newcopies = copies.copy()
                         for dest, source in pycompat.iteritems(childcopies):
                             prev = copies.get(source)
                             if prev is not None and prev[1] is not None:
                                 source = prev[1]
                             newcopies[dest] = (current_rev, source)
                         assert newcopies is not copies
                     if changes.removed:
                         for f in changes.removed:
                             if f in newcopies:
                                 if newcopies is copies:
                                     # copy on write to avoid affecting potential other
                                     # branches.  when there are no other branches, this
                                     # could be avoided.
                                     newcopies = copies.copy()
                                 newcopies[f] = (current_rev, None)
                 # check potential need to combine the data from another parent (for
                 # that child). See comment below for details.
                 if current_copies is None:
                     current_copies = newcopies
                 else:
                     # we are the second parent to work on c, we need to merge our
                     # work with the other.
                     #
                     # In case of conflict, parent 1 take precedence over parent 2.
                     # This is an arbitrary choice made anew when implementing
                     # changeset based copies. It was made without regards with
                     # potential filelog related behavior.
                     assert parent == 2
                     current_copies = _merge_copies_dict(
                         newcopies,
                         current_copies,
                         isancestor,
                         changes,
                         current_rev,
                     )
             all_copies[current_rev] = current_copies
 
         # filter out internal details and return a {dest: source mapping}
         final_copies = {}
         for dest, (tt, source) in all_copies[targetrev].items():
             if source is not None:
                 final_copies[dest] = source
     if not alwaysmatch:
         for filename in list(final_copies.keys()):
             if not match(filename):
                 del final_copies[filename]
     return final_copies
 
 
 # constant to decide which side to pick with _merge_copies_dict
 PICK_MINOR = 0
 PICK_MAJOR = 1
 PICK_EITHER = 2
 
 
 def _merge_copies_dict(minor, major, isancestor, changes, current_merge):
     """merge two copies-mapping together, minor and major
 
     In case of conflict, value from "major" will be picked.
 
     - `isancestors(low_rev, high_rev)`: callable return True if `low_rev` is an
                                         ancestors of `high_rev`,
 
     - `ismerged(path)`: callable return True if `path` have been merged in the
                         current revision,
 
     return the resulting dict (in practice, the "minor" object, updated)
     """
     for dest, value in major.items():
         other = minor.get(dest)
         if other is None:
             minor[dest] = value
         else:
             pick, overwrite = _compare_values(
                 changes, isancestor, dest, other, value
             )
             if overwrite:
                 if pick == PICK_MAJOR:
                     minor[dest] = (current_merge, value[1])
                 else:
                     minor[dest] = (current_merge, other[1])
             elif pick == PICK_MAJOR:
                 minor[dest] = value
     return minor
 
 
 def _compare_values(changes, isancestor, dest, minor, major):
     """compare two value within a _merge_copies_dict loop iteration
 
     return (pick, overwrite).
 
     - pick is one of PICK_MINOR, PICK_MAJOR or PICK_EITHER
     - overwrite is True if pick is a return of an ambiguity that needs resolution.
     """
     major_tt, major_value = major
     minor_tt, minor_value = minor
 
     if major_tt == minor_tt:
         # if it comes from the same revision it must be the same value
         assert major_value == minor_value
         return PICK_EITHER, False
     elif (
         changes is not None
         and minor_value is not None
         and major_value is None
         and dest in changes.salvaged
     ):
         # In this case, a deletion was reverted, the "alive" value overwrite
         # the deleted one.
         return PICK_MINOR, True
     elif (
         changes is not None
         and major_value is not None
         and minor_value is None
         and dest in changes.salvaged
     ):
         # In this case, a deletion was reverted, the "alive" value overwrite
         # the deleted one.
         return PICK_MAJOR, True
     elif isancestor(minor_tt, major_tt):
         if changes is not None and dest in changes.merged:
             # change to dest happened on the branch without copy-source change,
             # so both source are valid and "major" wins.
             return PICK_MAJOR, True
         else:
             return PICK_MAJOR, False
     elif isancestor(major_tt, minor_tt):
         if changes is not None and dest in changes.merged:
             # change to dest happened on the branch without copy-source change,
             # so both source are valid and "major" wins.
             return PICK_MAJOR, True
         else:
             return PICK_MINOR, False
     elif minor_value is None:
         # in case of conflict, the "alive" side wins.
         return PICK_MAJOR, True
     elif major_value is None:
         # in case of conflict, the "alive" side wins.
         return PICK_MINOR, True
     else:
         # in case of conflict where both side are alive, major wins.
         return PICK_MAJOR, True
 
 
 def _revinfo_getter_extra(repo):
     """return a function that return multiple data given a <rev>"i
 
     * p1: revision number of first parent
     * p2: revision number of first parent
     * p1copies: mapping of copies from p1
     * p2copies: mapping of copies from p2
     * removed: a list of removed files
     * ismerged: a callback to know if file was merged in that revision
     """
     cl = repo.changelog
     parents = cl.parentrevs
 
     def get_ismerged(rev):
         ctx = repo[rev]
 
         def ismerged(path):
             if path not in ctx.files():
                 return False
             fctx = ctx[path]
             parents = fctx._filelog.parents(fctx._filenode)
             nb_parents = 0
             for n in parents:
                 if n != nullid:
                     nb_parents += 1
             return nb_parents >= 2
 
         return ismerged
 
     def revinfo(rev):
         p1, p2 = parents(rev)
         ctx = repo[rev]
         p1copies, p2copies = ctx._copies
         removed = ctx.filesremoved()
         return p1, p2, p1copies, p2copies, removed, get_ismerged(rev)
 
     return revinfo
 
 
 def _combine_changeset_copies_extra(
     revs, children, targetrev, revinfo, match, isancestor
 ):
     """version of `_combine_changeset_copies` that works with the Google
     specific "extra" based storage for copy information"""
     all_copies = {}
     alwaysmatch = match.always()
     for r in revs:
         copies = all_copies.pop(r, None)
         if copies is None:
             # this is a root
             copies = {}
         for i, c in enumerate(children[r]):
             p1, p2, p1copies, p2copies, removed, ismerged = revinfo(c)
             if r == p1:
                 parent = 1
                 childcopies = p1copies
             else:
                 assert r == p2
                 parent = 2
                 childcopies = p2copies
             if not alwaysmatch:
                 childcopies = {
                     dst: src for dst, src in childcopies.items() if match(dst)
                 }
             newcopies = copies
             if childcopies:
                 newcopies = copies.copy()
                 for dest, source in pycompat.iteritems(childcopies):
                     prev = copies.get(source)
                     if prev is not None and prev[1] is not None:
                         source = prev[1]
                     newcopies[dest] = (c, source)
                 assert newcopies is not copies
             for f in removed:
                 if f in newcopies:
                     if newcopies is copies:
                         # copy on write to avoid affecting potential other
                         # branches.  when there are no other branches, this
                         # could be avoided.
                         newcopies = copies.copy()
                     newcopies[f] = (c, None)
             othercopies = all_copies.get(c)
             if othercopies is None:
                 all_copies[c] = newcopies
             else:
                 # we are the second parent to work on c, we need to merge our
                 # work with the other.
                 #
                 # In case of conflict, parent 1 take precedence over parent 2.
                 # This is an arbitrary choice made anew when implementing
                 # changeset based copies. It was made without regards with
                 # potential filelog related behavior.
                 if parent == 1:
                     _merge_copies_dict_extra(
                         othercopies, newcopies, isancestor, ismerged
                     )
                 else:
                     _merge_copies_dict_extra(
                         newcopies, othercopies, isancestor, ismerged
                     )
                     all_copies[c] = newcopies
 
     final_copies = {}
     for dest, (tt, source) in all_copies[targetrev].items():
         if source is not None:
             final_copies[dest] = source
     return final_copies
 
 
 def _merge_copies_dict_extra(minor, major, isancestor, ismerged):
     """version of `_merge_copies_dict` that works with the Google
     specific "extra" based storage for copy information"""
     for dest, value in major.items():
         other = minor.get(dest)
         if other is None:
             minor[dest] = value
         else:
             new_tt = value[0]
             other_tt = other[0]
             if value[1] == other[1]:
                 continue
             # content from "major" wins, unless it is older
             # than the branch point or there is a merge
             if (
                 new_tt == other_tt
                 or not isancestor(new_tt, other_tt)
                 or ismerged(dest)
             ):
                 minor[dest] = value
 
 
 def _forwardcopies(a, b, base=None, match=None):
     """find {dst@b: src@a} copy mapping where a is an ancestor of b"""
 
     if base is None:
         base = a
     match = a.repo().narrowmatch(match)
     # check for working copy
     if b.rev() is None:
         cm = _committedforwardcopies(a, b.p1(), base, match)
         # combine copies from dirstate if necessary
         copies = _chain(cm, _dirstatecopies(b._repo, match))
     else:
         copies = _committedforwardcopies(a, b, base, match)
     return copies
 
 
 def _backwardrenames(a, b, match):
     """find renames from a to b"""
     if a._repo.ui.config(b'experimental', b'copytrace') == b'off':
         return {}
 
     # We don't want to pass in "match" here, since that would filter
     # the destination by it. Since we're reversing the copies, we want
     # to filter the source instead.
     copies = _forwardcopies(b, a)
     return _reverse_renames(copies, a, match)
 
 
 def _reverse_renames(copies, dst, match):
     """given copies to context 'dst', finds renames from that context"""
     # Even though we're not taking copies into account, 1:n rename situations
     # can still exist (e.g. hg cp a b; hg mv a c). In those cases we
     # arbitrarily pick one of the renames.
     r = {}
     for k, v in sorted(pycompat.iteritems(copies)):
         if match and not match(v):
             continue
         # remove copies
         if v in dst:
             continue
         r[v] = k
     return r
 
 
 def pathcopies(x, y, match=None):
     """find {dst@y: src@x} copy mapping for directed compare"""
     repo = x._repo
     debug = repo.ui.debugflag and repo.ui.configbool(b'devel', b'debug.copies')
     if debug:
         repo.ui.debug(
             b'debug.copies: searching copies from %s to %s\n' % (x, y)
         )
     if x == y or not x or not y:
         return {}
     if y.rev() is None and x == y.p1():
         if debug:
             repo.ui.debug(b'debug.copies: search mode: dirstate\n')
         # short-circuit to avoid issues with merge states
         return _dirstatecopies(repo, match)
     a = y.ancestor(x)
     if a == x:
         if debug:
             repo.ui.debug(b'debug.copies: search mode: forward\n')
         copies = _forwardcopies(x, y, match=match)
     elif a == y:
         if debug:
             repo.ui.debug(b'debug.copies: search mode: backward\n')
         copies = _backwardrenames(x, y, match=match)
     else:
         if debug:
             repo.ui.debug(b'debug.copies: search mode: combined\n')
         base = None
         if a.rev() != nullrev:
             base = x
         x_copies = _forwardcopies(a, x)
         y_copies = _forwardcopies(a, y, base, match=match)
         same_keys = set(x_copies) & set(y_copies)
         for k in same_keys:
             if x_copies.get(k) == y_copies.get(k):
                 del x_copies[k]
                 del y_copies[k]
         x_backward_renames = _reverse_renames(x_copies, x, match)
         copies = _chain(
             x_backward_renames,
             y_copies,
         )
     _filter(x, y, copies)
     return copies
 
 
 def mergecopies(repo, c1, c2, base):
     """
     Finds moves and copies between context c1 and c2 that are relevant for
     merging. 'base' will be used as the merge base.
 
     Copytracing is used in commands like rebase, merge, unshelve, etc to merge
     files that were moved/ copied in one merge parent and modified in another.
     For example:
 
     o          ---> 4 another commit
     |
     |   o      ---> 3 commit that modifies a.txt
     |  /
     o /        ---> 2 commit that moves a.txt to b.txt
     |/
     o          ---> 1 merge base
 
     If we try to rebase revision 3 on revision 4, since there is no a.txt in
     revision 4, and if user have copytrace disabled, we prints the following
     message:
 
     ```other changed <file> which local deleted```
 
     Returns a tuple where:
 
     "branch_copies" an instance of branch_copies.
 
     "diverge" is a mapping of source name -> list of destination names
     for divergent renames.
 
     This function calls different copytracing algorithms based on config.
     """
     # avoid silly behavior for update from empty dir
     if not c1 or not c2 or c1 == c2:
         return branch_copies(), branch_copies(), {}
 
     narrowmatch = c1.repo().narrowmatch()
 
     # avoid silly behavior for parent -> working dir
     if c2.node() is None and c1.node() == repo.dirstate.p1():
         return (
             branch_copies(_dirstatecopies(repo, narrowmatch)),
             branch_copies(),
             {},
         )
 
     copytracing = repo.ui.config(b'experimental', b'copytrace')
     if stringutil.parsebool(copytracing) is False:
         # stringutil.parsebool() returns None when it is unable to parse the
         # value, so we should rely on making sure copytracing is on such cases
         return branch_copies(), branch_copies(), {}
 
     if usechangesetcentricalgo(repo):
         # The heuristics don't make sense when we need changeset-centric algos
         return _fullcopytracing(repo, c1, c2, base)
 
     # Copy trace disabling is explicitly below the node == p1 logic above
     # because the logic above is required for a simple copy to be kept across a
     # rebase.
     if copytracing == b'heuristics':
         # Do full copytracing if only non-public revisions are involved as
         # that will be fast enough and will also cover the copies which could
         # be missed by heuristics
         if _isfullcopytraceable(repo, c1, base):
             return _fullcopytracing(repo, c1, c2, base)
         return _heuristicscopytracing(repo, c1, c2, base)
     else:
         return _fullcopytracing(repo, c1, c2, base)
 
 
 def _isfullcopytraceable(repo, c1, base):
     """Checks that if base, source and destination are all no-public branches,
     if yes let's use the full copytrace algorithm for increased capabilities
     since it will be fast enough.
 
     `experimental.copytrace.sourcecommitlimit` can be used to set a limit for
     number of changesets from c1 to base such that if number of changesets are
     more than the limit, full copytracing algorithm won't be used.
     """
     if c1.rev() is None:
         c1 = c1.p1()
     if c1.mutable() and base.mutable():
         sourcecommitlimit = repo.ui.configint(
             b'experimental', b'copytrace.sourcecommitlimit'
         )
         commits = len(repo.revs(b'%d::%d', base.rev(), c1.rev()))
         return commits < sourcecommitlimit
     return False
 
 
 def _checksinglesidecopies(
     src, dsts1, m1, m2, mb, c2, base, copy, renamedelete
 ):
     if src not in m2:
         # deleted on side 2
         if src not in m1:
             # renamed on side 1, deleted on side 2
             renamedelete[src] = dsts1
     elif src not in mb:
         # Work around the "short-circuit to avoid issues with merge states"
         # thing in pathcopies(): pathcopies(x, y) can return a copy where the
         # destination doesn't exist in y.
         pass
     elif mb[src] != m2[src] and not _related(c2[src], base[src]):
         return
     elif mb[src] != m2[src] or mb.flags(src) != m2.flags(src):
         # modified on side 2
         for dst in dsts1:
             copy[dst] = src
 
 
 class branch_copies(object):
     """Information about copies made on one side of a merge/graft.
 
     "copy" is a mapping from destination name -> source name,
     where source is in c1 and destination is in c2 or vice-versa.
 
     "movewithdir" is a mapping from source name -> destination name,
     where the file at source present in one context but not the other
     needs to be moved to destination by the merge process, because the
     other context moved the directory it is in.
 
     "renamedelete" is a mapping of source name -> list of destination
     names for files deleted in c1 that were renamed in c2 or vice-versa.
 
     "dirmove" is a mapping of detected source dir -> destination dir renames.
     This is needed for handling changes to new files previously grafted into
     renamed directories.
     """
 
     def __init__(
         self, copy=None, renamedelete=None, dirmove=None, movewithdir=None
     ):
         self.copy = {} if copy is None else copy
         self.renamedelete = {} if renamedelete is None else renamedelete
         self.dirmove = {} if dirmove is None else dirmove
         self.movewithdir = {} if movewithdir is None else movewithdir
 
     def __repr__(self):
         return '<branch_copies\n  copy=%r\n  renamedelete=%r\n  dirmove=%r\n  movewithdir=%r\n>' % (
             self.copy,
             self.renamedelete,
             self.dirmove,
             self.movewithdir,
         )
 
 
 def _fullcopytracing(repo, c1, c2, base):
     """The full copytracing algorithm which finds all the new files that were
     added from merge base up to the top commit and for each file it checks if
     this file was copied from another file.
 
     This is pretty slow when a lot of changesets are involved but will track all
     the copies.
     """
     m1 = c1.manifest()
     m2 = c2.manifest()
     mb = base.manifest()
 
     copies1 = pathcopies(base, c1)
     copies2 = pathcopies(base, c2)
 
     if not (copies1 or copies2):
         return branch_copies(), branch_copies(), {}
 
     inversecopies1 = {}
     inversecopies2 = {}
     for dst, src in copies1.items():
         inversecopies1.setdefault(src, []).append(dst)
     for dst, src in copies2.items():
         inversecopies2.setdefault(src, []).append(dst)
 
     copy1 = {}
     copy2 = {}
     diverge = {}
     renamedelete1 = {}
     renamedelete2 = {}
     allsources = set(inversecopies1) | set(inversecopies2)
     for src in allsources:
         dsts1 = inversecopies1.get(src)
         dsts2 = inversecopies2.get(src)
         if dsts1 and dsts2:
             # copied/renamed on both sides
             if src not in m1 and src not in m2:
                 # renamed on both sides
                 dsts1 = set(dsts1)
                 dsts2 = set(dsts2)
                 # If there's some overlap in the rename destinations, we
                 # consider it not divergent. For example, if side 1 copies 'a'
                 # to 'b' and 'c' and deletes 'a', and side 2 copies 'a' to 'c'
                 # and 'd' and deletes 'a'.
                 if dsts1 & dsts2:
                     for dst in dsts1 & dsts2:
                         copy1[dst] = src
                         copy2[dst] = src
                 else:
                     diverge[src] = sorted(dsts1 | dsts2)
             elif src in m1 and src in m2:
                 # copied on both sides
                 dsts1 = set(dsts1)
                 dsts2 = set(dsts2)
                 for dst in dsts1 & dsts2:
                     copy1[dst] = src
                     copy2[dst] = src
             # TODO: Handle cases where it was renamed on one side and copied
             # on the other side
         elif dsts1:
             # copied/renamed only on side 1
             _checksinglesidecopies(
                 src, dsts1, m1, m2, mb, c2, base, copy1, renamedelete1
             )
         elif dsts2:
             # copied/renamed only on side 2
             _checksinglesidecopies(
                 src, dsts2, m2, m1, mb, c1, base, copy2, renamedelete2
             )
 
     # find interesting file sets from manifests
     cache = []
 
     def _get_addedfiles(idx):
         if not cache:
             addedinm1 = m1.filesnotin(mb, repo.narrowmatch())
             addedinm2 = m2.filesnotin(mb, repo.narrowmatch())
             u1 = sorted(addedinm1 - addedinm2)
             u2 = sorted(addedinm2 - addedinm1)
             cache.extend((u1, u2))
         return cache[idx]
 
     u1fn = lambda: _get_addedfiles(0)
     u2fn = lambda: _get_addedfiles(1)
     if repo.ui.debugflag:
         u1 = u1fn()
         u2 = u2fn()
 
         header = b"  unmatched files in %s"
         if u1:
             repo.ui.debug(
                 b"%s:\n   %s\n" % (header % b'local', b"\n   ".join(u1))
             )
         if u2:
             repo.ui.debug(
                 b"%s:\n   %s\n" % (header % b'other', b"\n   ".join(u2))
             )
 
         renamedeleteset = set()
         divergeset = set()
         for dsts in diverge.values():
             divergeset.update(dsts)
         for dsts in renamedelete1.values():
             renamedeleteset.update(dsts)
         for dsts in renamedelete2.values():
             renamedeleteset.update(dsts)
 
         repo.ui.debug(
             b"  all copies found (* = to merge, ! = divergent, "
             b"% = renamed and deleted):\n"
         )
         for side, copies in ((b"local", copies1), (b"remote", copies2)):
             if not copies:
                 continue
             repo.ui.debug(b"   on %s side:\n" % side)
             for f in sorted(copies):
                 note = b""
                 if f in copy1 or f in copy2:
                     note += b"*"
                 if f in divergeset:
                     note += b"!"
                 if f in renamedeleteset:
                     note += b"%"
                 repo.ui.debug(
                     b"    src: '%s' -> dst: '%s' %s\n" % (copies[f], f, note)
                 )
         del renamedeleteset
         del divergeset
 
     repo.ui.debug(b"  checking for directory renames\n")
 
     dirmove1, movewithdir2 = _dir_renames(repo, c1, copy1, copies1, u2fn)
     dirmove2, movewithdir1 = _dir_renames(repo, c2, copy2, copies2, u1fn)
 
     branch_copies1 = branch_copies(copy1, renamedelete1, dirmove1, movewithdir1)
     branch_copies2 = branch_copies(copy2, renamedelete2, dirmove2, movewithdir2)
 
     return branch_copies1, branch_copies2, diverge
 
 
 def _dir_renames(repo, ctx, copy, fullcopy, addedfilesfn):
     """Finds moved directories and files that should move with them.
 
     ctx: the context for one of the sides
     copy: files copied on the same side (as ctx)
     fullcopy: files copied on the same side (as ctx), including those that
               merge.manifestmerge() won't care about
     addedfilesfn: function returning added files on the other side (compared to
                   ctx)
     """
     # generate a directory move map
     invalid = set()
     dirmove = {}
 
     # examine each file copy for a potential directory move, which is
     # when all the files in a directory are moved to a new directory
     for dst, src in pycompat.iteritems(fullcopy):
         dsrc, ddst = pathutil.dirname(src), pathutil.dirname(dst)
         if dsrc in invalid:
             # already seen to be uninteresting
             continue
         elif ctx.hasdir(dsrc) and ctx.hasdir(ddst):
             # directory wasn't entirely moved locally
             invalid.add(dsrc)
         elif dsrc in dirmove and dirmove[dsrc] != ddst:
             # files from the same directory moved to two different places
             invalid.add(dsrc)
         else:
             # looks good so far
             dirmove[dsrc] = ddst
 
     for i in invalid:
         if i in dirmove:
             del dirmove[i]
     del invalid
 
     if not dirmove:
         return {}, {}
 
     dirmove = {k + b"/": v + b"/" for k, v in pycompat.iteritems(dirmove)}
 
     for d in dirmove:
         repo.ui.debug(
             b"   discovered dir src: '%s' -> dst: '%s'\n" % (d, dirmove[d])
         )
 
     # Sort the directories in reverse order, so we find children first
     # For example, if dir1/ was renamed to dir2/, and dir1/subdir1/
     # was renamed to dir2/subdir2/, we want to move dir1/subdir1/file
     # to dir2/subdir2/file (not dir2/subdir1/file)
     dirmove_children_first = sorted(dirmove, reverse=True)
 
     movewithdir = {}
     # check unaccounted nonoverlapping files against directory moves
     for f in addedfilesfn():
         if f not in fullcopy:
             for d in dirmove_children_first:
                 if f.startswith(d):
                     # new file added in a directory that was moved, move it
                     df = dirmove[d] + f[len(d) :]
                     if df not in copy:
                         movewithdir[f] = df
                         repo.ui.debug(
                             b"   pending file src: '%s' -> dst: '%s'\n"
                             % (f, df)
                         )
                     break
 
     return dirmove, movewithdir
 
 
 def _heuristicscopytracing(repo, c1, c2, base):
     """Fast copytracing using filename heuristics
 
     Assumes that moves or renames are of following two types:
 
     1) Inside a directory only (same directory name but different filenames)
     2) Move from one directory to another
                     (same filenames but different directory names)
 
     Works only when there are no merge commits in the "source branch".
     Source branch is commits from base up to c2 not including base.
 
     If merge is involved it fallbacks to _fullcopytracing().
 
     Can be used by setting the following config:
 
         [experimental]
         copytrace = heuristics
 
     In some cases the copy/move candidates found by heuristics can be very large
     in number and that will make the algorithm slow. The number of possible
     candidates to check can be limited by using the config
     `experimental.copytrace.movecandidateslimit` which defaults to 100.
     """
 
     if c1.rev() is None:
         c1 = c1.p1()
     if c2.rev() is None:
         c2 = c2.p1()
 
     changedfiles = set()
     m1 = c1.manifest()
     if not repo.revs(b'%d::%d', base.rev(), c2.rev()):
         # If base is not in c2 branch, we switch to fullcopytracing
         repo.ui.debug(
             b"switching to full copytracing as base is not "
             b"an ancestor of c2\n"
         )
         return _fullcopytracing(repo, c1, c2, base)
 
     ctx = c2
     while ctx != base:
         if len(ctx.parents()) == 2:
             # To keep things simple let's not handle merges
             repo.ui.debug(b"switching to full copytracing because of merges\n")
             return _fullcopytracing(repo, c1, c2, base)
         changedfiles.update(ctx.files())
         ctx = ctx.p1()
 
     copies2 = {}
     cp = _forwardcopies(base, c2)
     for dst, src in pycompat.iteritems(cp):
         if src in m1:
             copies2[dst] = src
 
     # file is missing if it isn't present in the destination, but is present in
     # the base and present in the source.
     # Presence in the base is important to exclude added files, presence in the
     # source is important to exclude removed files.
     filt = lambda f: f not in m1 and f in base and f in c2
     missingfiles = [f for f in changedfiles if filt(f)]
 
     copies1 = {}
     if missingfiles:
         basenametofilename = collections.defaultdict(list)
         dirnametofilename = collections.defaultdict(list)
 
         for f in m1.filesnotin(base.manifest()):
             basename = os.path.basename(f)
             dirname = os.path.dirname(f)
             basenametofilename[basename].append(f)
             dirnametofilename[dirname].append(f)
 
         for f in missingfiles:
             basename = os.path.basename(f)
             dirname = os.path.dirname(f)
             samebasename = basenametofilename[basename]
             samedirname = dirnametofilename[dirname]
             movecandidates = samebasename + samedirname
             # f is guaranteed to be present in c2, that's why
             # c2.filectx(f) won't fail
             f2 = c2.filectx(f)
             # we can have a lot of candidates which can slow down the heuristics
             # config value to limit the number of candidates moves to check
             maxcandidates = repo.ui.configint(
                 b'experimental', b'copytrace.movecandidateslimit'
             )
 
             if len(movecandidates) > maxcandidates:
                 repo.ui.status(
                     _(
                         b"skipping copytracing for '%s', more "
                         b"candidates than the limit: %d\n"
                     )
                     % (f, len(movecandidates))
                 )
                 continue
 
             for candidate in movecandidates:
                 f1 = c1.filectx(candidate)
                 if _related(f1, f2):
                     # if there are a few related copies then we'll merge
                     # changes into all of them. This matches the behaviour
                     # of upstream copytracing
                     copies1[candidate] = f
 
     return branch_copies(copies1), branch_copies(copies2), {}
 
 
 def _related(f1, f2):
     """return True if f1 and f2 filectx have a common ancestor
 
     Walk back to common ancestor to see if the two files originate
     from the same file. Since workingfilectx's rev() is None it messes
     up the integer comparison logic, hence the pre-step check for
     None (f1 and f2 can only be workingfilectx's initially).
     """
 
     if f1 == f2:
         return True  # a match
 
     g1, g2 = f1.ancestors(), f2.ancestors()
     try:
         f1r, f2r = f1.linkrev(), f2.linkrev()
 
         if f1r is None:
             f1 = next(g1)
         if f2r is None:
             f2 = next(g2)
 
         while True:
             f1r, f2r = f1.linkrev(), f2.linkrev()
             if f1r > f2r:
                 f1 = next(g1)
             elif f2r > f1r:
                 f2 = next(g2)
             else:  # f1 and f2 point to files in the same linkrev
                 return f1 == f2  # true if they point to the same file
     except StopIteration:
         return False
 
 
 def graftcopies(wctx, ctx, base):
     """reproduce copies between base and ctx in the wctx
 
     Unlike mergecopies(), this function will only consider copies between base
     and ctx; it will ignore copies between base and wctx. Also unlike
     mergecopies(), this function will apply copies to the working copy (instead
     of just returning information about the copies). That makes it cheaper
     (especially in the common case of base==ctx.p1()) and useful also when
     experimental.copytrace=off.
 
     merge.update() will have already marked most copies, but it will only
     mark copies if it thinks the source files are related (see
     merge._related()). It will also not mark copies if the file wasn't modified
     on the local side. This function adds the copies that were "missed"
     by merge.update().
     """
     new_copies = pathcopies(base, ctx)
     parent = wctx.p1()
     _filter(parent, wctx, new_copies)
     # Extra filtering to drop copy information for files that existed before
     # the graft. This is to handle the case of grafting a rename onto a commit
     # that already has the rename. Otherwise the presence of copy information
     # would result in the creation of an empty commit where we would prefer to
     # not create one.
     for dest, __ in list(new_copies.items()):
         if dest in parent:
             del new_copies[dest]
     for dst, src in pycompat.iteritems(new_copies):
         wctx[dst].markcopied(src)

mercurial/logcmdutil.py

@@ -1,1251 +1,1252 b''
 # logcmdutil.py - utility for log-like commands
 #
 # Copyright 2005-2007 Olivia Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import itertools
 import os
 import posixpath
 
 from .i18n import _
 from .node import (
     nullid,
+    nullrev,
     wdirid,
     wdirrev,
 )
 
 from .thirdparty import attr
 
 from . import (
     dagop,
     error,
     formatter,
     graphmod,
     match as matchmod,
     mdiff,
     merge,
     patch,
     pathutil,
     pycompat,
     revset,
     revsetlang,
     scmutil,
     smartset,
     templatekw,
     templater,
     util,
 )
 from .utils import (
     dateutil,
     stringutil,
 )
 
 
 if pycompat.TYPE_CHECKING:
     from typing import (
         Any,
         Callable,
         Dict,
         List,
         Optional,
         Sequence,
         Tuple,
     )
 
     for t in (Any, Callable, Dict, List, Optional, Tuple):
         assert t
 
 
 def getlimit(opts):
     """get the log limit according to option -l/--limit"""
     limit = opts.get(b'limit')
     if limit:
         try:
             limit = int(limit)
         except ValueError:
             raise error.Abort(_(b'limit must be a positive integer'))
         if limit <= 0:
             raise error.Abort(_(b'limit must be positive'))
     else:
         limit = None
     return limit
 
 
 def diff_parent(ctx):
     """get the context object to use as parent when diffing
 
 
     If diff.merge is enabled, an overlayworkingctx of the auto-merged parents will be returned.
     """
     repo = ctx.repo()
-    if repo.ui.configbool(b"diff", b"merge") and ctx.p2().node() != nullid:
+    if repo.ui.configbool(b"diff", b"merge") and ctx.p2().rev() != nullrev:
         # avoid cycle context -> subrepo -> cmdutil -> logcmdutil
         from . import context
 
         wctx = context.overlayworkingctx(repo)
         wctx.setbase(ctx.p1())
         with repo.ui.configoverride(
             {
                 (
                     b"ui",
                     b"forcemerge",
                 ): b"internal:merge3-lie-about-conflicts",
             },
             b"merge-diff",
         ):
             repo.ui.pushbuffer()
             merge.merge(ctx.p2(), wc=wctx)
             repo.ui.popbuffer()
         return wctx
     else:
         return ctx.p1()
 
 
 def diffordiffstat(
     ui,
     repo,
     diffopts,
     ctx1,
     ctx2,
     match,
     changes=None,
     stat=False,
     fp=None,
     graphwidth=0,
     prefix=b'',
     root=b'',
     listsubrepos=False,
     hunksfilterfn=None,
 ):
     '''show diff or diffstat.'''
     if root:
         relroot = pathutil.canonpath(repo.root, repo.getcwd(), root)
     else:
         relroot = b''
     copysourcematch = None
 
     def compose(f, g):
         return lambda x: f(g(x))
 
     def pathfn(f):
         return posixpath.join(prefix, f)
 
     if relroot != b'':
         # XXX relative roots currently don't work if the root is within a
         # subrepo
         uipathfn = scmutil.getuipathfn(repo, legacyrelativevalue=True)
         uirelroot = uipathfn(pathfn(relroot))
         relroot += b'/'
         for matchroot in match.files():
             if not matchroot.startswith(relroot):
                 ui.warn(
                     _(b'warning: %s not inside relative root %s\n')
                     % (uipathfn(pathfn(matchroot)), uirelroot)
                 )
 
         relrootmatch = scmutil.match(ctx2, pats=[relroot], default=b'path')
         match = matchmod.intersectmatchers(match, relrootmatch)
         copysourcematch = relrootmatch
 
         checkroot = repo.ui.configbool(
             b'devel', b'all-warnings'
         ) or repo.ui.configbool(b'devel', b'check-relroot')
 
         def relrootpathfn(f):
             if checkroot and not f.startswith(relroot):
                 raise AssertionError(
                     b"file %s doesn't start with relroot %s" % (f, relroot)
                 )
             return f[len(relroot) :]
 
         pathfn = compose(relrootpathfn, pathfn)
 
     if stat:
         diffopts = diffopts.copy(context=0, noprefix=False)
         width = 80
         if not ui.plain():
             width = ui.termwidth() - graphwidth
         # If an explicit --root was given, don't respect ui.relative-paths
         if not relroot:
             pathfn = compose(scmutil.getuipathfn(repo), pathfn)
 
     chunks = ctx2.diff(
         ctx1,
         match,
         changes,
         opts=diffopts,
         pathfn=pathfn,
         copysourcematch=copysourcematch,
         hunksfilterfn=hunksfilterfn,
     )
 
     if fp is not None or ui.canwritewithoutlabels():
         out = fp or ui
         if stat:
             chunks = [patch.diffstat(util.iterlines(chunks), width=width)]
         for chunk in util.filechunkiter(util.chunkbuffer(chunks)):
             out.write(chunk)
     else:
         if stat:
             chunks = patch.diffstatui(util.iterlines(chunks), width=width)
         else:
             chunks = patch.difflabel(
                 lambda chunks, **kwargs: chunks, chunks, opts=diffopts
             )
         if ui.canbatchlabeledwrites():
 
             def gen():
                 for chunk, label in chunks:
                     yield ui.label(chunk, label=label)
 
             for chunk in util.filechunkiter(util.chunkbuffer(gen())):
                 ui.write(chunk)
         else:
             for chunk, label in chunks:
                 ui.write(chunk, label=label)
 
     node2 = ctx2.node()
     for subpath, sub in scmutil.itersubrepos(ctx1, ctx2):
         tempnode2 = node2
         try:
             if node2 is not None:
                 tempnode2 = ctx2.substate[subpath][1]
         except KeyError:
             # A subrepo that existed in node1 was deleted between node1 and
             # node2 (inclusive). Thus, ctx2's substate won't contain that
             # subpath. The best we can do is to ignore it.
             tempnode2 = None
         submatch = matchmod.subdirmatcher(subpath, match)
         subprefix = repo.wvfs.reljoin(prefix, subpath)
         if listsubrepos or match.exact(subpath) or any(submatch.files()):
             sub.diff(
                 ui,
                 diffopts,
                 tempnode2,
                 submatch,
                 changes=changes,
                 stat=stat,
                 fp=fp,
                 prefix=subprefix,
             )
 
 
 class changesetdiffer(object):
     """Generate diff of changeset with pre-configured filtering functions"""
 
     def _makefilematcher(self, ctx):
         return scmutil.matchall(ctx.repo())
 
     def _makehunksfilter(self, ctx):
         return None
 
     def showdiff(self, ui, ctx, diffopts, graphwidth=0, stat=False):
         diffordiffstat(
             ui,
             ctx.repo(),
             diffopts,
             diff_parent(ctx),
             ctx,
             match=self._makefilematcher(ctx),
             stat=stat,
             graphwidth=graphwidth,
             hunksfilterfn=self._makehunksfilter(ctx),
         )
 
 
 def changesetlabels(ctx):
     labels = [b'log.changeset', b'changeset.%s' % ctx.phasestr()]
     if ctx.obsolete():
         labels.append(b'changeset.obsolete')
     if ctx.isunstable():
         labels.append(b'changeset.unstable')
         for instability in ctx.instabilities():
             labels.append(b'instability.%s' % instability)
     return b' '.join(labels)
 
 
 class changesetprinter(object):
     '''show changeset information when templating not requested.'''
 
     def __init__(self, ui, repo, differ=None, diffopts=None, buffered=False):
         self.ui = ui
         self.repo = repo
         self.buffered = buffered
         self._differ = differ or changesetdiffer()
         self._diffopts = patch.diffallopts(ui, diffopts)
         self._includestat = diffopts and diffopts.get(b'stat')
         self._includediff = diffopts and diffopts.get(b'patch')
         self.header = {}
         self.hunk = {}
         self.lastheader = None
         self.footer = None
         self._columns = templatekw.getlogcolumns()
 
     def flush(self, ctx):
         rev = ctx.rev()
         if rev in self.header:
             h = self.header[rev]
             if h != self.lastheader:
                 self.lastheader = h
                 self.ui.write(h)
             del self.header[rev]
         if rev in self.hunk:
             self.ui.write(self.hunk[rev])
             del self.hunk[rev]
 
     def close(self):
         if self.footer:
             self.ui.write(self.footer)
 
     def show(self, ctx, copies=None, **props):
         props = pycompat.byteskwargs(props)
         if self.buffered:
             self.ui.pushbuffer(labeled=True)
             self._show(ctx, copies, props)
             self.hunk[ctx.rev()] = self.ui.popbuffer()
         else:
             self._show(ctx, copies, props)
 
     def _show(self, ctx, copies, props):
         '''show a single changeset or file revision'''
         changenode = ctx.node()
         graphwidth = props.get(b'graphwidth', 0)
 
         if self.ui.quiet:
             self.ui.write(
                 b"%s\n" % scmutil.formatchangeid(ctx), label=b'log.node'
             )
             return
 
         columns = self._columns
         self.ui.write(
             columns[b'changeset'] % scmutil.formatchangeid(ctx),
             label=changesetlabels(ctx),
         )
 
         # branches are shown first before any other names due to backwards
         # compatibility
         branch = ctx.branch()
         # don't show the default branch name
         if branch != b'default':
             self.ui.write(columns[b'branch'] % branch, label=b'log.branch')
 
         for nsname, ns in pycompat.iteritems(self.repo.names):
             # branches has special logic already handled above, so here we just
             # skip it
             if nsname == b'branches':
                 continue
             # we will use the templatename as the color name since those two
             # should be the same
             for name in ns.names(self.repo, changenode):
                 self.ui.write(ns.logfmt % name, label=b'log.%s' % ns.colorname)
         if self.ui.debugflag:
             self.ui.write(
                 columns[b'phase'] % ctx.phasestr(), label=b'log.phase'
             )
         for pctx in scmutil.meaningfulparents(self.repo, ctx):
             label = b'log.parent changeset.%s' % pctx.phasestr()
             self.ui.write(
                 columns[b'parent'] % scmutil.formatchangeid(pctx), label=label
             )
 
         if self.ui.debugflag:
             mnode = ctx.manifestnode()
             if mnode is None:
                 mnode = wdirid
                 mrev = wdirrev
             else:
                 mrev = self.repo.manifestlog.rev(mnode)
             self.ui.write(
                 columns[b'manifest']
                 % scmutil.formatrevnode(self.ui, mrev, mnode),
                 label=b'ui.debug log.manifest',
             )
         self.ui.write(columns[b'user'] % ctx.user(), label=b'log.user')
         self.ui.write(
             columns[b'date'] % dateutil.datestr(ctx.date()), label=b'log.date'
         )
 
         if ctx.isunstable():
             instabilities = ctx.instabilities()
             self.ui.write(
                 columns[b'instability'] % b', '.join(instabilities),
                 label=b'log.instability',
             )
 
         elif ctx.obsolete():
             self._showobsfate(ctx)
 
         self._exthook(ctx)
 
         if self.ui.debugflag:
             files = ctx.p1().status(ctx)
             for key, value in zip(
                 [b'files', b'files+', b'files-'],
                 [files.modified, files.added, files.removed],
             ):
                 if value:
                     self.ui.write(
                         columns[key] % b" ".join(value),
                         label=b'ui.debug log.files',
                     )
         elif ctx.files() and self.ui.verbose:
             self.ui.write(
                 columns[b'files'] % b" ".join(ctx.files()),
                 label=b'ui.note log.files',
             )
         if copies and self.ui.verbose:
             copies = [b'%s (%s)' % c for c in copies]
             self.ui.write(
                 columns[b'copies'] % b' '.join(copies),
                 label=b'ui.note log.copies',
             )
 
         extra = ctx.extra()
         if extra and self.ui.debugflag:
             for key, value in sorted(extra.items()):
                 self.ui.write(
                     columns[b'extra'] % (key, stringutil.escapestr(value)),
                     label=b'ui.debug log.extra',
                 )
 
         description = ctx.description().strip()
         if description:
             if self.ui.verbose:
                 self.ui.write(
                     _(b"description:\n"), label=b'ui.note log.description'
                 )
                 self.ui.write(description, label=b'ui.note log.description')
                 self.ui.write(b"\n\n")
             else:
                 self.ui.write(
                     columns[b'summary'] % description.splitlines()[0],
                     label=b'log.summary',
                 )
         self.ui.write(b"\n")
 
         self._showpatch(ctx, graphwidth)
 
     def _showobsfate(self, ctx):
         # TODO: do not depend on templater
         tres = formatter.templateresources(self.repo.ui, self.repo)
         t = formatter.maketemplater(
             self.repo.ui,
             b'{join(obsfate, "\n")}',
             defaults=templatekw.keywords,
             resources=tres,
         )
         obsfate = t.renderdefault({b'ctx': ctx}).splitlines()
 
         if obsfate:
             for obsfateline in obsfate:
                 self.ui.write(
                     self._columns[b'obsolete'] % obsfateline,
                     label=b'log.obsfate',
                 )
 
     def _exthook(self, ctx):
         """empty method used by extension as a hook point"""
 
     def _showpatch(self, ctx, graphwidth=0):
         if self._includestat:
             self._differ.showdiff(
                 self.ui, ctx, self._diffopts, graphwidth, stat=True
             )
         if self._includestat and self._includediff:
             self.ui.write(b"\n")
         if self._includediff:
             self._differ.showdiff(
                 self.ui, ctx, self._diffopts, graphwidth, stat=False
             )
         if self._includestat or self._includediff:
             self.ui.write(b"\n")
 
 
 class changesetformatter(changesetprinter):
     """Format changeset information by generic formatter"""
 
     def __init__(
         self, ui, repo, fm, differ=None, diffopts=None, buffered=False
     ):
         changesetprinter.__init__(self, ui, repo, differ, diffopts, buffered)
         self._diffopts = patch.difffeatureopts(ui, diffopts, git=True)
         self._fm = fm
 
     def close(self):
         self._fm.end()
 
     def _show(self, ctx, copies, props):
         '''show a single changeset or file revision'''
         fm = self._fm
         fm.startitem()
         fm.context(ctx=ctx)
         fm.data(rev=scmutil.intrev(ctx), node=fm.hexfunc(scmutil.binnode(ctx)))
 
         datahint = fm.datahint()
         if self.ui.quiet and not datahint:
             return
 
         fm.data(
             branch=ctx.branch(),
             phase=ctx.phasestr(),
             user=ctx.user(),
             date=fm.formatdate(ctx.date()),
             desc=ctx.description(),
             bookmarks=fm.formatlist(ctx.bookmarks(), name=b'bookmark'),
             tags=fm.formatlist(ctx.tags(), name=b'tag'),
             parents=fm.formatlist(
                 [fm.hexfunc(c.node()) for c in ctx.parents()], name=b'node'
             ),
         )
 
         if self.ui.debugflag or b'manifest' in datahint:
             fm.data(manifest=fm.hexfunc(ctx.manifestnode() or wdirid))
         if self.ui.debugflag or b'extra' in datahint:
             fm.data(extra=fm.formatdict(ctx.extra()))
 
         if (
             self.ui.debugflag
             or b'modified' in datahint
             or b'added' in datahint
             or b'removed' in datahint
         ):
             files = ctx.p1().status(ctx)
             fm.data(
                 modified=fm.formatlist(files.modified, name=b'file'),
                 added=fm.formatlist(files.added, name=b'file'),
                 removed=fm.formatlist(files.removed, name=b'file'),
             )
 
         verbose = not self.ui.debugflag and self.ui.verbose
         if verbose or b'files' in datahint:
             fm.data(files=fm.formatlist(ctx.files(), name=b'file'))
         if verbose and copies or b'copies' in datahint:
             fm.data(
                 copies=fm.formatdict(copies or {}, key=b'name', value=b'source')
             )
 
         if self._includestat or b'diffstat' in datahint:
             self.ui.pushbuffer()
             self._differ.showdiff(self.ui, ctx, self._diffopts, stat=True)
             fm.data(diffstat=self.ui.popbuffer())
         if self._includediff or b'diff' in datahint:
             self.ui.pushbuffer()
             self._differ.showdiff(self.ui, ctx, self._diffopts, stat=False)
             fm.data(diff=self.ui.popbuffer())
 
 
 class changesettemplater(changesetprinter):
     """format changeset information.
 
     Note: there are a variety of convenience functions to build a
     changesettemplater for common cases. See functions such as:
     maketemplater, changesetdisplayer, buildcommittemplate, or other
     functions that use changesest_templater.
     """
 
     # Arguments before "buffered" used to be positional. Consider not
     # adding/removing arguments before "buffered" to not break callers.
     def __init__(
         self, ui, repo, tmplspec, differ=None, diffopts=None, buffered=False
     ):
         changesetprinter.__init__(self, ui, repo, differ, diffopts, buffered)
         # tres is shared with _graphnodeformatter()
         self._tresources = tres = formatter.templateresources(ui, repo)
         self.t = formatter.loadtemplater(
             ui,
             tmplspec,
             defaults=templatekw.keywords,
             resources=tres,
             cache=templatekw.defaulttempl,
         )
         self._counter = itertools.count()
 
         self._tref = tmplspec.ref
         self._parts = {
             b'header': b'',
             b'footer': b'',
             tmplspec.ref: tmplspec.ref,
             b'docheader': b'',
             b'docfooter': b'',
             b'separator': b'',
         }
         if tmplspec.mapfile:
             # find correct templates for current mode, for backward
             # compatibility with 'log -v/-q/--debug' using a mapfile
             tmplmodes = [
                 (True, b''),
                 (self.ui.verbose, b'_verbose'),
                 (self.ui.quiet, b'_quiet'),
                 (self.ui.debugflag, b'_debug'),
             ]
             for mode, postfix in tmplmodes:
                 for t in self._parts:
                     cur = t + postfix
                     if mode and cur in self.t:
                         self._parts[t] = cur
         else:
             partnames = [p for p in self._parts.keys() if p != tmplspec.ref]
             m = formatter.templatepartsmap(tmplspec, self.t, partnames)
             self._parts.update(m)
 
         if self._parts[b'docheader']:
             self.ui.write(self.t.render(self._parts[b'docheader'], {}))
 
     def close(self):
         if self._parts[b'docfooter']:
             if not self.footer:
                 self.footer = b""
             self.footer += self.t.render(self._parts[b'docfooter'], {})
         return super(changesettemplater, self).close()
 
     def _show(self, ctx, copies, props):
         '''show a single changeset or file revision'''
         props = props.copy()
         props[b'ctx'] = ctx
         props[b'index'] = index = next(self._counter)
         props[b'revcache'] = {b'copies': copies}
         graphwidth = props.get(b'graphwidth', 0)
 
         # write separator, which wouldn't work well with the header part below
         # since there's inherently a conflict between header (across items) and
         # separator (per item)
         if self._parts[b'separator'] and index > 0:
             self.ui.write(self.t.render(self._parts[b'separator'], {}))
 
         # write header
         if self._parts[b'header']:
             h = self.t.render(self._parts[b'header'], props)
             if self.buffered:
                 self.header[ctx.rev()] = h
             else:
                 if self.lastheader != h:
                     self.lastheader = h
                     self.ui.write(h)
 
         # write changeset metadata, then patch if requested
         key = self._parts[self._tref]
         self.ui.write(self.t.render(key, props))
         self._exthook(ctx)
         self._showpatch(ctx, graphwidth)
 
         if self._parts[b'footer']:
             if not self.footer:
                 self.footer = self.t.render(self._parts[b'footer'], props)
 
 
 def templatespec(tmpl, mapfile):
     assert not (tmpl and mapfile)
     if mapfile:
         return formatter.mapfile_templatespec(b'changeset', mapfile)
     else:
         return formatter.literal_templatespec(tmpl)
 
 
 def _lookuptemplate(ui, tmpl, style):
     """Find the template matching the given template spec or style
 
     See formatter.lookuptemplate() for details.
     """
 
     # ui settings
     if not tmpl and not style:  # template are stronger than style
         tmpl = ui.config(b'command-templates', b'log')
         if tmpl:
             return formatter.literal_templatespec(templater.unquotestring(tmpl))
         else:
             style = util.expandpath(ui.config(b'ui', b'style'))
 
     if not tmpl and style:
         mapfile = style
         fp = None
         if not os.path.split(mapfile)[0]:
             (mapname, fp) = templater.try_open_template(
                 b'map-cmdline.' + mapfile
             ) or templater.try_open_template(mapfile)
             if mapname:
                 mapfile = mapname
         return formatter.mapfile_templatespec(b'changeset', mapfile, fp)
 
     return formatter.lookuptemplate(ui, b'changeset', tmpl)
 
 
 def maketemplater(ui, repo, tmpl, buffered=False):
     """Create a changesettemplater from a literal template 'tmpl'
     byte-string."""
     spec = formatter.literal_templatespec(tmpl)
     return changesettemplater(ui, repo, spec, buffered=buffered)
 
 
 def changesetdisplayer(ui, repo, opts, differ=None, buffered=False):
     """show one changeset using template or regular display.
 
     Display format will be the first non-empty hit of:
     1. option 'template'
     2. option 'style'
     3. [command-templates] setting 'log'
     4. [ui] setting 'style'
     If all of these values are either the unset or the empty string,
     regular display via changesetprinter() is done.
     """
     postargs = (differ, opts, buffered)
     spec = _lookuptemplate(ui, opts.get(b'template'), opts.get(b'style'))
 
     # machine-readable formats have slightly different keyword set than
     # plain templates, which are handled by changesetformatter.
     # note that {b'pickle', b'debug'} can also be added to the list if needed.
     if spec.ref in {b'cbor', b'json'}:
         fm = ui.formatter(b'log', opts)
         return changesetformatter(ui, repo, fm, *postargs)
 
     if not spec.ref and not spec.tmpl and not spec.mapfile:
         return changesetprinter(ui, repo, *postargs)
 
     return changesettemplater(ui, repo, spec, *postargs)
 
 
 @attr.s
 class walkopts(object):
     """Options to configure a set of revisions and file matcher factory
     to scan revision/file history
     """
 
     # raw command-line parameters, which a matcher will be built from
     pats = attr.ib()  # type: List[bytes]
     opts = attr.ib()  # type: Dict[bytes, Any]
 
     # a list of revset expressions to be traversed; if follow, it specifies
     # the start revisions
     revspec = attr.ib()  # type: List[bytes]
 
     # miscellaneous queries to filter revisions (see "hg help log" for details)
     bookmarks = attr.ib(default=attr.Factory(list))  # type: List[bytes]
     branches = attr.ib(default=attr.Factory(list))  # type: List[bytes]
     date = attr.ib(default=None)  # type: Optional[bytes]
     keywords = attr.ib(default=attr.Factory(list))  # type: List[bytes]
     no_merges = attr.ib(default=False)  # type: bool
     only_merges = attr.ib(default=False)  # type: bool
     prune_ancestors = attr.ib(default=attr.Factory(list))  # type: List[bytes]
     users = attr.ib(default=attr.Factory(list))  # type: List[bytes]
 
     # miscellaneous matcher arguments
     include_pats = attr.ib(default=attr.Factory(list))  # type: List[bytes]
     exclude_pats = attr.ib(default=attr.Factory(list))  # type: List[bytes]
 
     # 0: no follow, 1: follow first, 2: follow both parents
     follow = attr.ib(default=0)  # type: int
 
     # do not attempt filelog-based traversal, which may be fast but cannot
     # include revisions where files were removed
     force_changelog_traversal = attr.ib(default=False)  # type: bool
 
     # filter revisions by file patterns, which should be disabled only if
     # you want to include revisions where files were unmodified
     filter_revisions_by_pats = attr.ib(default=True)  # type: bool
 
     # sort revisions prior to traversal: 'desc', 'topo', or None
     sort_revisions = attr.ib(default=None)  # type: Optional[bytes]
 
     # limit number of changes displayed; None means unlimited
     limit = attr.ib(default=None)  # type: Optional[int]
 
 
 def parseopts(ui, pats, opts):
     # type: (Any, Sequence[bytes], Dict[bytes, Any]) -> walkopts
     """Parse log command options into walkopts
 
     The returned walkopts will be passed in to getrevs() or makewalker().
     """
     if opts.get(b'follow_first'):
         follow = 1
     elif opts.get(b'follow'):
         follow = 2
     else:
         follow = 0
 
     if opts.get(b'graph'):
         if ui.configbool(b'experimental', b'log.topo'):
             sort_revisions = b'topo'
         else:
             sort_revisions = b'desc'
     else:
         sort_revisions = None
 
     return walkopts(
         pats=pats,
         opts=opts,
         revspec=opts.get(b'rev', []),
         bookmarks=opts.get(b'bookmark', []),
         # branch and only_branch are really aliases and must be handled at
         # the same time
         branches=opts.get(b'branch', []) + opts.get(b'only_branch', []),
         date=opts.get(b'date'),
         keywords=opts.get(b'keyword', []),
         no_merges=bool(opts.get(b'no_merges')),
         only_merges=bool(opts.get(b'only_merges')),
         prune_ancestors=opts.get(b'prune', []),
         users=opts.get(b'user', []),
         include_pats=opts.get(b'include', []),
         exclude_pats=opts.get(b'exclude', []),
         follow=follow,
         force_changelog_traversal=bool(opts.get(b'removed')),
         sort_revisions=sort_revisions,
         limit=getlimit(opts),
     )
 
 
 def _makematcher(repo, revs, wopts):
     """Build matcher and expanded patterns from log options
 
     If --follow, revs are the revisions to follow from.
 
     Returns (match, pats, slowpath) where
     - match: a matcher built from the given pats and -I/-X opts
     - pats: patterns used (globs are expanded on Windows)
     - slowpath: True if patterns aren't as simple as scanning filelogs
     """
     # pats/include/exclude are passed to match.match() directly in
     # _matchfiles() revset, but a log-like command should build its matcher
     # with scmutil.match(). The difference is input pats are globbed on
     # platforms without shell expansion (windows).
     wctx = repo[None]
     match, pats = scmutil.matchandpats(wctx, wopts.pats, wopts.opts)
     slowpath = match.anypats() or (
         not match.always() and wopts.force_changelog_traversal
     )
     if not slowpath:
         if wopts.follow and wopts.revspec:
             # There may be the case that a path doesn't exist in some (but
             # not all) of the specified start revisions, but let's consider
             # the path is valid. Missing files will be warned by the matcher.
             startctxs = [repo[r] for r in revs]
             for f in match.files():
                 found = False
                 for c in startctxs:
                     if f in c:
                         found = True
                     elif c.hasdir(f):
                         # If a directory exists in any of the start revisions,
                         # take the slow path.
                         found = slowpath = True
                 if not found:
                     raise error.Abort(
                         _(
                             b'cannot follow file not in any of the specified '
                             b'revisions: "%s"'
                         )
                         % f
                     )
         elif wopts.follow:
             for f in match.files():
                 if f not in wctx:
                     # If the file exists, it may be a directory, so let it
                     # take the slow path.
                     if os.path.exists(repo.wjoin(f)):
                         slowpath = True
                         continue
                     else:
                         raise error.Abort(
                             _(
                                 b'cannot follow file not in parent '
                                 b'revision: "%s"'
                             )
                             % f
                         )
                 filelog = repo.file(f)
                 if not filelog:
                     # A file exists in wdir but not in history, which means
                     # the file isn't committed yet.
                     raise error.Abort(
                         _(b'cannot follow nonexistent file: "%s"') % f
                     )
         else:
             for f in match.files():
                 filelog = repo.file(f)
                 if not filelog:
                     # A zero count may be a directory or deleted file, so
                     # try to find matching entries on the slow path.
                     slowpath = True
 
         # We decided to fall back to the slowpath because at least one
         # of the paths was not a file. Check to see if at least one of them
         # existed in history - in that case, we'll continue down the
         # slowpath; otherwise, we can turn off the slowpath
         if slowpath:
             for path in match.files():
                 if not path or path in repo.store:
                     break
             else:
                 slowpath = False
 
     return match, pats, slowpath
 
 
 def _fileancestors(repo, revs, match, followfirst):
     fctxs = []
     for r in revs:
         ctx = repo[r]
         fctxs.extend(ctx[f].introfilectx() for f in ctx.walk(match))
 
     # When displaying a revision with --patch --follow FILE, we have
     # to know which file of the revision must be diffed. With
     # --follow, we want the names of the ancestors of FILE in the
     # revision, stored in "fcache". "fcache" is populated as a side effect
     # of the graph traversal.
     fcache = {}
 
     def filematcher(ctx):
         return scmutil.matchfiles(repo, fcache.get(scmutil.intrev(ctx), []))
 
     def revgen():
         for rev, cs in dagop.filectxancestors(fctxs, followfirst=followfirst):
             fcache[rev] = [c.path() for c in cs]
             yield rev
 
     return smartset.generatorset(revgen(), iterasc=False), filematcher
 
 
 def _makenofollowfilematcher(repo, pats, opts):
     '''hook for extensions to override the filematcher for non-follow cases'''
     return None
 
 
 _opt2logrevset = {
     b'no_merges': (b'not merge()', None),
     b'only_merges': (b'merge()', None),
     b'_matchfiles': (None, b'_matchfiles(%ps)'),
     b'date': (b'date(%s)', None),
     b'branch': (b'branch(%s)', b'%lr'),
     b'_patslog': (b'filelog(%s)', b'%lr'),
     b'keyword': (b'keyword(%s)', b'%lr'),
     b'prune': (b'ancestors(%s)', b'not %lr'),
     b'user': (b'user(%s)', b'%lr'),
 }
 
 
 def _makerevset(repo, wopts, slowpath):
     """Return a revset string built from log options and file patterns"""
     opts = {
         b'branch': [b'literal:' + repo.lookupbranch(b) for b in wopts.branches],
         b'date': wopts.date,
         b'keyword': wopts.keywords,
         b'no_merges': wopts.no_merges,
         b'only_merges': wopts.only_merges,
         b'prune': wopts.prune_ancestors,
         b'user': [b'literal:' + v for v in wopts.users],
     }
 
     if wopts.filter_revisions_by_pats and slowpath:
         # pats/include/exclude cannot be represented as separate
         # revset expressions as their filtering logic applies at file
         # level. For instance "-I a -X b" matches a revision touching
         # "a" and "b" while "file(a) and not file(b)" does
         # not. Besides, filesets are evaluated against the working
         # directory.
         matchargs = [b'r:', b'd:relpath']
         for p in wopts.pats:
             matchargs.append(b'p:' + p)
         for p in wopts.include_pats:
             matchargs.append(b'i:' + p)
         for p in wopts.exclude_pats:
             matchargs.append(b'x:' + p)
         opts[b'_matchfiles'] = matchargs
     elif wopts.filter_revisions_by_pats and not wopts.follow:
         opts[b'_patslog'] = list(wopts.pats)
 
     expr = []
     for op, val in sorted(pycompat.iteritems(opts)):
         if not val:
             continue
         revop, listop = _opt2logrevset[op]
         if revop and b'%' not in revop:
             expr.append(revop)
         elif not listop:
             expr.append(revsetlang.formatspec(revop, val))
         else:
             if revop:
                 val = [revsetlang.formatspec(revop, v) for v in val]
             expr.append(revsetlang.formatspec(listop, val))
 
     if wopts.bookmarks:
         expr.append(
             revsetlang.formatspec(
                 b'%lr',
                 [scmutil.format_bookmark_revspec(v) for v in wopts.bookmarks],
             )
         )
 
     if expr:
         expr = b'(' + b' and '.join(expr) + b')'
     else:
         expr = None
     return expr
 
 
 def _initialrevs(repo, wopts):
     """Return the initial set of revisions to be filtered or followed"""
     if wopts.revspec:
         revs = scmutil.revrange(repo, wopts.revspec)
     elif wopts.follow and repo.dirstate.p1() == nullid:
         revs = smartset.baseset()
     elif wopts.follow:
         revs = repo.revs(b'.')
     else:
         revs = smartset.spanset(repo)
         revs.reverse()
     return revs
 
 
 def makewalker(repo, wopts):
     # type: (Any, walkopts) -> Tuple[smartset.abstractsmartset, Optional[Callable[[Any], matchmod.basematcher]]]
     """Build (revs, makefilematcher) to scan revision/file history
 
     - revs is the smartset to be traversed.
     - makefilematcher is a function to map ctx to a matcher for that revision
     """
     revs = _initialrevs(repo, wopts)
     if not revs:
         return smartset.baseset(), None
     # TODO: might want to merge slowpath with wopts.force_changelog_traversal
     match, pats, slowpath = _makematcher(repo, revs, wopts)
     wopts = attr.evolve(wopts, pats=pats)
 
     filematcher = None
     if wopts.follow:
         if slowpath or match.always():
             revs = dagop.revancestors(repo, revs, followfirst=wopts.follow == 1)
         else:
             assert not wopts.force_changelog_traversal
             revs, filematcher = _fileancestors(
                 repo, revs, match, followfirst=wopts.follow == 1
             )
         revs.reverse()
     if filematcher is None:
         filematcher = _makenofollowfilematcher(repo, wopts.pats, wopts.opts)
     if filematcher is None:
 
         def filematcher(ctx):
             return match
 
     expr = _makerevset(repo, wopts, slowpath)
     if wopts.sort_revisions:
         assert wopts.sort_revisions in {b'topo', b'desc'}
         if wopts.sort_revisions == b'topo':
             if not revs.istopo():
                 revs = dagop.toposort(revs, repo.changelog.parentrevs)
                 # TODO: try to iterate the set lazily
                 revs = revset.baseset(list(revs), istopo=True)
         elif not (revs.isdescending() or revs.istopo()):
             # User-specified revs might be unsorted
             revs.sort(reverse=True)
     if expr:
         matcher = revset.match(None, expr)
         revs = matcher(repo, revs)
     if wopts.limit is not None:
         revs = revs.slice(0, wopts.limit)
 
     return revs, filematcher
 
 
 def getrevs(repo, wopts):
     # type: (Any, walkopts) -> Tuple[smartset.abstractsmartset, Optional[changesetdiffer]]
     """Return (revs, differ) where revs is a smartset
 
     differ is a changesetdiffer with pre-configured file matcher.
     """
     revs, filematcher = makewalker(repo, wopts)
     if not revs:
         return revs, None
     differ = changesetdiffer()
     differ._makefilematcher = filematcher
     return revs, differ
 
 
 def _parselinerangeopt(repo, opts):
     """Parse --line-range log option and return a list of tuples (filename,
     (fromline, toline)).
     """
     linerangebyfname = []
     for pat in opts.get(b'line_range', []):
         try:
             pat, linerange = pat.rsplit(b',', 1)
         except ValueError:
             raise error.Abort(_(b'malformatted line-range pattern %s') % pat)
         try:
             fromline, toline = map(int, linerange.split(b':'))
         except ValueError:
             raise error.Abort(_(b"invalid line range for %s") % pat)
         msg = _(b"line range pattern '%s' must match exactly one file") % pat
         fname = scmutil.parsefollowlinespattern(repo, None, pat, msg)
         linerangebyfname.append(
             (fname, util.processlinerange(fromline, toline))
         )
     return linerangebyfname
 
 
 def getlinerangerevs(repo, userrevs, opts):
     """Return (revs, differ).
 
     "revs" are revisions obtained by processing "line-range" log options and
     walking block ancestors of each specified file/line-range.
 
     "differ" is a changesetdiffer with pre-configured file matcher and hunks
     filter.
     """
     wctx = repo[None]
 
     # Two-levels map of "rev -> file ctx -> [line range]".
     linerangesbyrev = {}
     for fname, (fromline, toline) in _parselinerangeopt(repo, opts):
         if fname not in wctx:
             raise error.Abort(
                 _(b'cannot follow file not in parent revision: "%s"') % fname
             )
         fctx = wctx.filectx(fname)
         for fctx, linerange in dagop.blockancestors(fctx, fromline, toline):
             rev = fctx.introrev()
             if rev is None:
                 rev = wdirrev
             if rev not in userrevs:
                 continue
             linerangesbyrev.setdefault(rev, {}).setdefault(
                 fctx.path(), []
             ).append(linerange)
 
     def nofilterhunksfn(fctx, hunks):
         return hunks
 
     def hunksfilter(ctx):
         fctxlineranges = linerangesbyrev.get(scmutil.intrev(ctx))
         if fctxlineranges is None:
             return nofilterhunksfn
 
         def filterfn(fctx, hunks):
             lineranges = fctxlineranges.get(fctx.path())
             if lineranges is not None:
                 for hr, lines in hunks:
                     if hr is None:  # binary
                         yield hr, lines
                         continue
                     if any(mdiff.hunkinrange(hr[2:], lr) for lr in lineranges):
                         yield hr, lines
             else:
                 for hunk in hunks:
                     yield hunk
 
         return filterfn
 
     def filematcher(ctx):
         files = list(linerangesbyrev.get(scmutil.intrev(ctx), []))
         return scmutil.matchfiles(repo, files)
 
     revs = sorted(linerangesbyrev, reverse=True)
 
     differ = changesetdiffer()
     differ._makefilematcher = filematcher
     differ._makehunksfilter = hunksfilter
     return smartset.baseset(revs), differ
 
 
 def _graphnodeformatter(ui, displayer):
     spec = ui.config(b'command-templates', b'graphnode')
     if not spec:
         return templatekw.getgraphnode  # fast path for "{graphnode}"
 
     spec = templater.unquotestring(spec)
     if isinstance(displayer, changesettemplater):
         # reuse cache of slow templates
         tres = displayer._tresources
     else:
         tres = formatter.templateresources(ui)
     templ = formatter.maketemplater(
         ui, spec, defaults=templatekw.keywords, resources=tres
     )
 
     def formatnode(repo, ctx, cache):
         props = {b'ctx': ctx, b'repo': repo}
         return templ.renderdefault(props)
 
     return formatnode
 
 
 def displaygraph(ui, repo, dag, displayer, edgefn, getcopies=None, props=None):
     props = props or {}
     formatnode = _graphnodeformatter(ui, displayer)
     state = graphmod.asciistate()
     styles = state.styles
 
     # only set graph styling if HGPLAIN is not set.
     if ui.plain(b'graph'):
         # set all edge styles to |, the default pre-3.8 behaviour
         styles.update(dict.fromkeys(styles, b'|'))
     else:
         edgetypes = {
             b'parent': graphmod.PARENT,
             b'grandparent': graphmod.GRANDPARENT,
             b'missing': graphmod.MISSINGPARENT,
         }
         for name, key in edgetypes.items():
             # experimental config: experimental.graphstyle.*
             styles[key] = ui.config(
                 b'experimental', b'graphstyle.%s' % name, styles[key]
             )
             if not styles[key]:
                 styles[key] = None
 
         # experimental config: experimental.graphshorten
         state.graphshorten = ui.configbool(b'experimental', b'graphshorten')
 
     formatnode_cache = {}
     for rev, type, ctx, parents in dag:
         char = formatnode(repo, ctx, formatnode_cache)
         copies = getcopies(ctx) if getcopies else None
         edges = edgefn(type, char, state, rev, parents)
         firstedge = next(edges)
         width = firstedge[2]
         displayer.show(
             ctx, copies=copies, graphwidth=width, **pycompat.strkwargs(props)
         )
         lines = displayer.hunk.pop(rev).split(b'\n')
         if not lines[-1]:
             del lines[-1]
         displayer.flush(ctx)
         for type, char, width, coldata in itertools.chain([firstedge], edges):
             graphmod.ascii(ui, state, type, char, lines, coldata)
             lines = []
     displayer.close()
 
 
 def displaygraphrevs(ui, repo, revs, displayer, getrenamed):
     revdag = graphmod.dagwalker(repo, revs)
     displaygraph(ui, repo, revdag, displayer, graphmod.asciiedges, getrenamed)
 
 
 def displayrevs(ui, repo, revs, displayer, getcopies):
     for rev in revs:
         ctx = repo[rev]
         copies = getcopies(ctx) if getcopies else None
         displayer.show(ctx, copies=copies)
         displayer.flush(ctx)
     displayer.close()
 
 
 def checkunsupportedgraphflags(pats, opts):
     for op in [b"newest_first"]:
         if op in opts and opts[op]:
             raise error.Abort(
                 _(b"-G/--graph option is incompatible with --%s")
                 % op.replace(b"_", b"-")
             )
 
 
 def graphrevs(repo, nodes, opts):
     limit = getlimit(opts)
     nodes.reverse()
     if limit is not None:
         nodes = nodes[:limit]
     return graphmod.nodes(repo, nodes)

mercurial/mergestate.py

@@ -1,840 +1,841 b''
 from __future__ import absolute_import
 
 import collections
 import errno
 import shutil
 import struct
 
 from .i18n import _
 from .node import (
     bin,
     hex,
     nullhex,
     nullid,
+    nullrev,
 )
 from . import (
     error,
     filemerge,
     pycompat,
     util,
 )
 from .utils import hashutil
 
 _pack = struct.pack
 _unpack = struct.unpack
 
 
 def _droponode(data):
     # used for compatibility for v1
     bits = data.split(b'\0')
     bits = bits[:-2] + bits[-1:]
     return b'\0'.join(bits)
 
 
 def _filectxorabsent(hexnode, ctx, f):
     if hexnode == nullhex:
         return filemerge.absentfilectx(ctx, f)
     else:
         return ctx[f]
 
 
 # Merge state record types. See ``mergestate`` docs for more.
 
 ####
 # merge records which records metadata about a current merge
 # exists only once in a mergestate
 #####
 RECORD_LOCAL = b'L'
 RECORD_OTHER = b'O'
 # record merge labels
 RECORD_LABELS = b'l'
 
 #####
 # record extra information about files, with one entry containing info about one
 # file. Hence, multiple of them can exists
 #####
 RECORD_FILE_VALUES = b'f'
 
 #####
 # merge records which represents state of individual merges of files/folders
 # These are top level records for each entry containing merge related info.
 # Each record of these has info about one file. Hence multiple of them can
 # exists
 #####
 RECORD_MERGED = b'F'
 RECORD_CHANGEDELETE_CONFLICT = b'C'
 # the path was dir on one side of merge and file on another
 RECORD_PATH_CONFLICT = b'P'
 
 #####
 # possible state which a merge entry can have. These are stored inside top-level
 # merge records mentioned just above.
 #####
 MERGE_RECORD_UNRESOLVED = b'u'
 MERGE_RECORD_RESOLVED = b'r'
 MERGE_RECORD_UNRESOLVED_PATH = b'pu'
 MERGE_RECORD_RESOLVED_PATH = b'pr'
 # represents that the file was automatically merged in favor
 # of other version. This info is used on commit.
 # This is now deprecated and commit related information is now
 # stored in RECORD_FILE_VALUES
 MERGE_RECORD_MERGED_OTHER = b'o'
 
 #####
 # top level record which stores other unknown records. Multiple of these can
 # exists
 #####
 RECORD_OVERRIDE = b't'
 
 #####
 # legacy records which are no longer used but kept to prevent breaking BC
 #####
 # This record was release in 5.4 and usage was removed in 5.5
 LEGACY_RECORD_RESOLVED_OTHER = b'R'
 # This record was release in 3.7 and usage was removed in 5.6
 LEGACY_RECORD_DRIVER_RESOLVED = b'd'
 # This record was release in 3.7 and usage was removed in 5.6
 LEGACY_MERGE_DRIVER_STATE = b'm'
 # This record was release in 3.7 and usage was removed in 5.6
 LEGACY_MERGE_DRIVER_MERGE = b'D'
 
 
 ACTION_FORGET = b'f'
 ACTION_REMOVE = b'r'
 ACTION_ADD = b'a'
 ACTION_GET = b'g'
 ACTION_PATH_CONFLICT = b'p'
 ACTION_PATH_CONFLICT_RESOLVE = b'pr'
 ACTION_ADD_MODIFIED = b'am'
 ACTION_CREATED = b'c'
 ACTION_DELETED_CHANGED = b'dc'
 ACTION_CHANGED_DELETED = b'cd'
 ACTION_MERGE = b'm'
 ACTION_LOCAL_DIR_RENAME_GET = b'dg'
 ACTION_DIR_RENAME_MOVE_LOCAL = b'dm'
 ACTION_KEEP = b'k'
 # the file was absent on local side before merge and we should
 # keep it absent (absent means file not present, it can be a result
 # of file deletion, rename etc.)
 ACTION_KEEP_ABSENT = b'ka'
 # the file is absent on the ancestor and remote side of the merge
 # hence this file is new and we should keep it
 ACTION_KEEP_NEW = b'kn'
 ACTION_EXEC = b'e'
 ACTION_CREATED_MERGE = b'cm'
 
 # actions which are no op
 NO_OP_ACTIONS = (
     ACTION_KEEP,
     ACTION_KEEP_ABSENT,
     ACTION_KEEP_NEW,
 )
 
 
 class _mergestate_base(object):
     """track 3-way merge state of individual files
 
     The merge state is stored on disk when needed. Two files are used: one with
     an old format (version 1), and one with a new format (version 2). Version 2
     stores a superset of the data in version 1, including new kinds of records
     in the future. For more about the new format, see the documentation for
     `_readrecordsv2`.
 
     Each record can contain arbitrary content, and has an associated type. This
     `type` should be a letter. If `type` is uppercase, the record is mandatory:
     versions of Mercurial that don't support it should abort. If `type` is
     lowercase, the record can be safely ignored.
 
     Currently known records:
 
     L: the node of the "local" part of the merge (hexified version)
     O: the node of the "other" part of the merge (hexified version)
     F: a file to be merged entry
     C: a change/delete or delete/change conflict
     P: a path conflict (file vs directory)
     f: a (filename, dictionary) tuple of optional values for a given file
     l: the labels for the parts of the merge.
 
     Merge record states (stored in self._state, indexed by filename):
     u: unresolved conflict
     r: resolved conflict
     pu: unresolved path conflict (file conflicts with directory)
     pr: resolved path conflict
     o: file was merged in favor of other parent of merge (DEPRECATED)
 
     The resolve command transitions between 'u' and 'r' for conflicts and
     'pu' and 'pr' for path conflicts.
     """
 
     def __init__(self, repo):
         """Initialize the merge state.
 
         Do not use this directly! Instead call read() or clean()."""
         self._repo = repo
         self._state = {}
         self._stateextras = collections.defaultdict(dict)
         self._local = None
         self._other = None
         self._labels = None
         # contains a mapping of form:
         # {filename : (merge_return_value, action_to_be_performed}
         # these are results of re-running merge process
         # this dict is used to perform actions on dirstate caused by re-running
         # the merge
         self._results = {}
         self._dirty = False
 
     def reset(self):
         pass
 
     def start(self, node, other, labels=None):
         self._local = node
         self._other = other
         self._labels = labels
 
     @util.propertycache
     def local(self):
         if self._local is None:
             msg = b"local accessed but self._local isn't set"
             raise error.ProgrammingError(msg)
         return self._local
 
     @util.propertycache
     def localctx(self):
         return self._repo[self.local]
 
     @util.propertycache
     def other(self):
         if self._other is None:
             msg = b"other accessed but self._other isn't set"
             raise error.ProgrammingError(msg)
         return self._other
 
     @util.propertycache
     def otherctx(self):
         return self._repo[self.other]
 
     def active(self):
         """Whether mergestate is active.
 
         Returns True if there appears to be mergestate. This is a rough proxy
         for "is a merge in progress."
         """
         return bool(self._local) or bool(self._state)
 
     def commit(self):
         """Write current state on disk (if necessary)"""
 
     @staticmethod
     def getlocalkey(path):
         """hash the path of a local file context for storage in the .hg/merge
         directory."""
 
         return hex(hashutil.sha1(path).digest())
 
     def _make_backup(self, fctx, localkey):
         raise NotImplementedError()
 
     def _restore_backup(self, fctx, localkey, flags):
         raise NotImplementedError()
 
     def add(self, fcl, fco, fca, fd):
         """add a new (potentially?) conflicting file the merge state
         fcl: file context for local,
         fco: file context for remote,
         fca: file context for ancestors,
         fd:  file path of the resulting merge.
 
         note: also write the local version to the `.hg/merge` directory.
         """
         if fcl.isabsent():
             localkey = nullhex
         else:
             localkey = mergestate.getlocalkey(fcl.path())
             self._make_backup(fcl, localkey)
         self._state[fd] = [
             MERGE_RECORD_UNRESOLVED,
             localkey,
             fcl.path(),
             fca.path(),
             hex(fca.filenode()),
             fco.path(),
             hex(fco.filenode()),
             fcl.flags(),
         ]
         self._stateextras[fd][b'ancestorlinknode'] = hex(fca.node())
         self._dirty = True
 
     def addpathconflict(self, path, frename, forigin):
         """add a new conflicting path to the merge state
         path:    the path that conflicts
         frename: the filename the conflicting file was renamed to
         forigin: origin of the file ('l' or 'r' for local/remote)
         """
         self._state[path] = [MERGE_RECORD_UNRESOLVED_PATH, frename, forigin]
         self._dirty = True
 
     def addcommitinfo(self, path, data):
         """stores information which is required at commit
         into _stateextras"""
         self._stateextras[path].update(data)
         self._dirty = True
 
     def __contains__(self, dfile):
         return dfile in self._state
 
     def __getitem__(self, dfile):
         return self._state[dfile][0]
 
     def __iter__(self):
         return iter(sorted(self._state))
 
     def files(self):
         return self._state.keys()
 
     def mark(self, dfile, state):
         self._state[dfile][0] = state
         self._dirty = True
 
     def unresolved(self):
         """Obtain the paths of unresolved files."""
 
         for f, entry in pycompat.iteritems(self._state):
             if entry[0] in (
                 MERGE_RECORD_UNRESOLVED,
                 MERGE_RECORD_UNRESOLVED_PATH,
             ):
                 yield f
 
     def allextras(self):
         """ return all extras information stored with the mergestate """
         return self._stateextras
 
     def extras(self, filename):
         """ return extras stored with the mergestate for the given filename """
         return self._stateextras[filename]
 
     def _resolve(self, preresolve, dfile, wctx):
         """rerun merge process for file path `dfile`.
         Returns whether the merge was completed and the return value of merge
         obtained from filemerge._filemerge().
         """
         if self[dfile] in (
             MERGE_RECORD_RESOLVED,
             LEGACY_RECORD_DRIVER_RESOLVED,
         ):
             return True, 0
         stateentry = self._state[dfile]
         state, localkey, lfile, afile, anode, ofile, onode, flags = stateentry
         octx = self._repo[self._other]
         extras = self.extras(dfile)
         anccommitnode = extras.get(b'ancestorlinknode')
         if anccommitnode:
             actx = self._repo[anccommitnode]
         else:
             actx = None
         fcd = _filectxorabsent(localkey, wctx, dfile)
         fco = _filectxorabsent(onode, octx, ofile)
         # TODO: move this to filectxorabsent
         fca = self._repo.filectx(afile, fileid=anode, changectx=actx)
         # "premerge" x flags
         flo = fco.flags()
         fla = fca.flags()
         if b'x' in flags + flo + fla and b'l' not in flags + flo + fla:
-            if fca.node() == nullid and flags != flo:
+            if fca.rev() == nullrev and flags != flo:
                 if preresolve:
                     self._repo.ui.warn(
                         _(
                             b'warning: cannot merge flags for %s '
                             b'without common ancestor - keeping local flags\n'
                         )
                         % afile
                     )
             elif flags == fla:
                 flags = flo
         if preresolve:
             # restore local
             if localkey != nullhex:
                 self._restore_backup(wctx[dfile], localkey, flags)
             else:
                 wctx[dfile].remove(ignoremissing=True)
             complete, merge_ret, deleted = filemerge.premerge(
                 self._repo,
                 wctx,
                 self._local,
                 lfile,
                 fcd,
                 fco,
                 fca,
                 labels=self._labels,
             )
         else:
             complete, merge_ret, deleted = filemerge.filemerge(
                 self._repo,
                 wctx,
                 self._local,
                 lfile,
                 fcd,
                 fco,
                 fca,
                 labels=self._labels,
             )
         if merge_ret is None:
             # If return value of merge is None, then there are no real conflict
             del self._state[dfile]
             self._dirty = True
         elif not merge_ret:
             self.mark(dfile, MERGE_RECORD_RESOLVED)
 
         if complete:
             action = None
             if deleted:
                 if fcd.isabsent():
                     # dc: local picked. Need to drop if present, which may
                     # happen on re-resolves.
                     action = ACTION_FORGET
                 else:
                     # cd: remote picked (or otherwise deleted)
                     action = ACTION_REMOVE
             else:
                 if fcd.isabsent():  # dc: remote picked
                     action = ACTION_GET
                 elif fco.isabsent():  # cd: local picked
                     if dfile in self.localctx:
                         action = ACTION_ADD_MODIFIED
                     else:
                         action = ACTION_ADD
                 # else: regular merges (no action necessary)
             self._results[dfile] = merge_ret, action
 
         return complete, merge_ret
 
     def preresolve(self, dfile, wctx):
         """run premerge process for dfile
 
         Returns whether the merge is complete, and the exit code."""
         return self._resolve(True, dfile, wctx)
 
     def resolve(self, dfile, wctx):
         """run merge process (assuming premerge was run) for dfile
 
         Returns the exit code of the merge."""
         return self._resolve(False, dfile, wctx)[1]
 
     def counts(self):
         """return counts for updated, merged and removed files in this
         session"""
         updated, merged, removed = 0, 0, 0
         for r, action in pycompat.itervalues(self._results):
             if r is None:
                 updated += 1
             elif r == 0:
                 if action == ACTION_REMOVE:
                     removed += 1
                 else:
                     merged += 1
         return updated, merged, removed
 
     def unresolvedcount(self):
         """get unresolved count for this merge (persistent)"""
         return len(list(self.unresolved()))
 
     def actions(self):
         """return lists of actions to perform on the dirstate"""
         actions = {
             ACTION_REMOVE: [],
             ACTION_FORGET: [],
             ACTION_ADD: [],
             ACTION_ADD_MODIFIED: [],
             ACTION_GET: [],
         }
         for f, (r, action) in pycompat.iteritems(self._results):
             if action is not None:
                 actions[action].append((f, None, b"merge result"))
         return actions
 
 
 class mergestate(_mergestate_base):
 
     statepathv1 = b'merge/state'
     statepathv2 = b'merge/state2'
 
     @staticmethod
     def clean(repo):
         """Initialize a brand new merge state, removing any existing state on
         disk."""
         ms = mergestate(repo)
         ms.reset()
         return ms
 
     @staticmethod
     def read(repo):
         """Initialize the merge state, reading it from disk."""
         ms = mergestate(repo)
         ms._read()
         return ms
 
     def _read(self):
         """Analyse each record content to restore a serialized state from disk
 
         This function process "record" entry produced by the de-serialization
         of on disk file.
         """
         unsupported = set()
         records = self._readrecords()
         for rtype, record in records:
             if rtype == RECORD_LOCAL:
                 self._local = bin(record)
             elif rtype == RECORD_OTHER:
                 self._other = bin(record)
             elif rtype == LEGACY_MERGE_DRIVER_STATE:
                 pass
             elif rtype in (
                 RECORD_MERGED,
                 RECORD_CHANGEDELETE_CONFLICT,
                 RECORD_PATH_CONFLICT,
                 LEGACY_MERGE_DRIVER_MERGE,
                 LEGACY_RECORD_RESOLVED_OTHER,
             ):
                 bits = record.split(b'\0')
                 # merge entry type MERGE_RECORD_MERGED_OTHER is deprecated
                 # and we now store related information in _stateextras, so
                 # lets write to _stateextras directly
                 if bits[1] == MERGE_RECORD_MERGED_OTHER:
                     self._stateextras[bits[0]][b'filenode-source'] = b'other'
                 else:
                     self._state[bits[0]] = bits[1:]
             elif rtype == RECORD_FILE_VALUES:
                 filename, rawextras = record.split(b'\0', 1)
                 extraparts = rawextras.split(b'\0')
                 extras = {}
                 i = 0
                 while i < len(extraparts):
                     extras[extraparts[i]] = extraparts[i + 1]
                     i += 2
 
                 self._stateextras[filename] = extras
             elif rtype == RECORD_LABELS:
                 labels = record.split(b'\0', 2)
                 self._labels = [l for l in labels if len(l) > 0]
             elif not rtype.islower():
                 unsupported.add(rtype)
 
         if unsupported:
             raise error.UnsupportedMergeRecords(unsupported)
 
     def _readrecords(self):
         """Read merge state from disk and return a list of record (TYPE, data)
 
         We read data from both v1 and v2 files and decide which one to use.
 
         V1 has been used by version prior to 2.9.1 and contains less data than
         v2. We read both versions and check if no data in v2 contradicts
         v1. If there is not contradiction we can safely assume that both v1
         and v2 were written at the same time and use the extract data in v2. If
         there is contradiction we ignore v2 content as we assume an old version
         of Mercurial has overwritten the mergestate file and left an old v2
         file around.
 
         returns list of record [(TYPE, data), ...]"""
         v1records = self._readrecordsv1()
         v2records = self._readrecordsv2()
         if self._v1v2match(v1records, v2records):
             return v2records
         else:
             # v1 file is newer than v2 file, use it
             # we have to infer the "other" changeset of the merge
             # we cannot do better than that with v1 of the format
             mctx = self._repo[None].parents()[-1]
             v1records.append((RECORD_OTHER, mctx.hex()))
             # add place holder "other" file node information
             # nobody is using it yet so we do no need to fetch the data
             # if mctx was wrong `mctx[bits[-2]]` may fails.
             for idx, r in enumerate(v1records):
                 if r[0] == RECORD_MERGED:
                     bits = r[1].split(b'\0')
                     bits.insert(-2, b'')
                     v1records[idx] = (r[0], b'\0'.join(bits))
             return v1records
 
     def _v1v2match(self, v1records, v2records):
         oldv2 = set()  # old format version of v2 record
         for rec in v2records:
             if rec[0] == RECORD_LOCAL:
                 oldv2.add(rec)
             elif rec[0] == RECORD_MERGED:
                 # drop the onode data (not contained in v1)
                 oldv2.add((RECORD_MERGED, _droponode(rec[1])))
         for rec in v1records:
             if rec not in oldv2:
                 return False
         else:
             return True
 
     def _readrecordsv1(self):
         """read on disk merge state for version 1 file
 
         returns list of record [(TYPE, data), ...]
 
         Note: the "F" data from this file are one entry short
               (no "other file node" entry)
         """
         records = []
         try:
             f = self._repo.vfs(self.statepathv1)
             for i, l in enumerate(f):
                 if i == 0:
                     records.append((RECORD_LOCAL, l[:-1]))
                 else:
                     records.append((RECORD_MERGED, l[:-1]))
             f.close()
         except IOError as err:
             if err.errno != errno.ENOENT:
                 raise
         return records
 
     def _readrecordsv2(self):
         """read on disk merge state for version 2 file
 
         This format is a list of arbitrary records of the form:
 
           [type][length][content]
 
         `type` is a single character, `length` is a 4 byte integer, and
         `content` is an arbitrary byte sequence of length `length`.
 
         Mercurial versions prior to 3.7 have a bug where if there are
         unsupported mandatory merge records, attempting to clear out the merge
         state with hg update --clean or similar aborts. The 't' record type
         works around that by writing out what those versions treat as an
         advisory record, but later versions interpret as special: the first
         character is the 'real' record type and everything onwards is the data.
 
         Returns list of records [(TYPE, data), ...]."""
         records = []
         try:
             f = self._repo.vfs(self.statepathv2)
             data = f.read()
             off = 0
             end = len(data)
             while off < end:
                 rtype = data[off : off + 1]
                 off += 1
                 length = _unpack(b'>I', data[off : (off + 4)])[0]
                 off += 4
                 record = data[off : (off + length)]
                 off += length
                 if rtype == RECORD_OVERRIDE:
                     rtype, record = record[0:1], record[1:]
                 records.append((rtype, record))
             f.close()
         except IOError as err:
             if err.errno != errno.ENOENT:
                 raise
         return records
 
     def commit(self):
         if self._dirty:
             records = self._makerecords()
             self._writerecords(records)
             self._dirty = False
 
     def _makerecords(self):
         records = []
         records.append((RECORD_LOCAL, hex(self._local)))
         records.append((RECORD_OTHER, hex(self._other)))
         # Write out state items. In all cases, the value of the state map entry
         # is written as the contents of the record. The record type depends on
         # the type of state that is stored, and capital-letter records are used
         # to prevent older versions of Mercurial that do not support the feature
         # from loading them.
         for filename, v in pycompat.iteritems(self._state):
             if v[0] in (
                 MERGE_RECORD_UNRESOLVED_PATH,
                 MERGE_RECORD_RESOLVED_PATH,
             ):
                 # Path conflicts. These are stored in 'P' records.  The current
                 # resolution state ('pu' or 'pr') is stored within the record.
                 records.append(
                     (RECORD_PATH_CONFLICT, b'\0'.join([filename] + v))
                 )
             elif v[1] == nullhex or v[6] == nullhex:
                 # Change/Delete or Delete/Change conflicts. These are stored in
                 # 'C' records. v[1] is the local file, and is nullhex when the
                 # file is deleted locally ('dc'). v[6] is the remote file, and
                 # is nullhex when the file is deleted remotely ('cd').
                 records.append(
                     (RECORD_CHANGEDELETE_CONFLICT, b'\0'.join([filename] + v))
                 )
             else:
                 # Normal files.  These are stored in 'F' records.
                 records.append((RECORD_MERGED, b'\0'.join([filename] + v)))
         for filename, extras in sorted(pycompat.iteritems(self._stateextras)):
             rawextras = b'\0'.join(
                 b'%s\0%s' % (k, v) for k, v in pycompat.iteritems(extras)
             )
             records.append(
                 (RECORD_FILE_VALUES, b'%s\0%s' % (filename, rawextras))
             )
         if self._labels is not None:
             labels = b'\0'.join(self._labels)
             records.append((RECORD_LABELS, labels))
         return records
 
     def _writerecords(self, records):
         """Write current state on disk (both v1 and v2)"""
         self._writerecordsv1(records)
         self._writerecordsv2(records)
 
     def _writerecordsv1(self, records):
         """Write current state on disk in a version 1 file"""
         f = self._repo.vfs(self.statepathv1, b'wb')
         irecords = iter(records)
         lrecords = next(irecords)
         assert lrecords[0] == RECORD_LOCAL
         f.write(hex(self._local) + b'\n')
         for rtype, data in irecords:
             if rtype == RECORD_MERGED:
                 f.write(b'%s\n' % _droponode(data))
         f.close()
 
     def _writerecordsv2(self, records):
         """Write current state on disk in a version 2 file
 
         See the docstring for _readrecordsv2 for why we use 't'."""
         # these are the records that all version 2 clients can read
         allowlist = (RECORD_LOCAL, RECORD_OTHER, RECORD_MERGED)
         f = self._repo.vfs(self.statepathv2, b'wb')
         for key, data in records:
             assert len(key) == 1
             if key not in allowlist:
                 key, data = RECORD_OVERRIDE, b'%s%s' % (key, data)
             format = b'>sI%is' % len(data)
             f.write(_pack(format, key, len(data), data))
         f.close()
 
     def _make_backup(self, fctx, localkey):
         self._repo.vfs.write(b'merge/' + localkey, fctx.data())
 
     def _restore_backup(self, fctx, localkey, flags):
         with self._repo.vfs(b'merge/' + localkey) as f:
             fctx.write(f.read(), flags)
 
     def reset(self):
         shutil.rmtree(self._repo.vfs.join(b'merge'), True)
 
 
 class memmergestate(_mergestate_base):
     def __init__(self, repo):
         super(memmergestate, self).__init__(repo)
         self._backups = {}
 
     def _make_backup(self, fctx, localkey):
         self._backups[localkey] = fctx.data()
 
     def _restore_backup(self, fctx, localkey, flags):
         fctx.write(self._backups[localkey], flags)
 
 
 def recordupdates(repo, actions, branchmerge, getfiledata):
     """record merge actions to the dirstate"""
     # remove (must come first)
     for f, args, msg in actions.get(ACTION_REMOVE, []):
         if branchmerge:
             repo.dirstate.remove(f)
         else:
             repo.dirstate.drop(f)
 
     # forget (must come first)
     for f, args, msg in actions.get(ACTION_FORGET, []):
         repo.dirstate.drop(f)
 
     # resolve path conflicts
     for f, args, msg in actions.get(ACTION_PATH_CONFLICT_RESOLVE, []):
         (f0, origf0) = args
         repo.dirstate.add(f)
         repo.dirstate.copy(origf0, f)
         if f0 == origf0:
             repo.dirstate.remove(f0)
         else:
             repo.dirstate.drop(f0)
 
     # re-add
     for f, args, msg in actions.get(ACTION_ADD, []):
         repo.dirstate.add(f)
 
     # re-add/mark as modified
     for f, args, msg in actions.get(ACTION_ADD_MODIFIED, []):
         if branchmerge:
             repo.dirstate.normallookup(f)
         else:
             repo.dirstate.add(f)
 
     # exec change
     for f, args, msg in actions.get(ACTION_EXEC, []):
         repo.dirstate.normallookup(f)
 
     # keep
     for f, args, msg in actions.get(ACTION_KEEP, []):
         pass
 
     # keep deleted
     for f, args, msg in actions.get(ACTION_KEEP_ABSENT, []):
         pass
 
     # keep new
     for f, args, msg in actions.get(ACTION_KEEP_NEW, []):
         pass
 
     # get
     for f, args, msg in actions.get(ACTION_GET, []):
         if branchmerge:
             repo.dirstate.otherparent(f)
         else:
             parentfiledata = getfiledata[f] if getfiledata else None
             repo.dirstate.normal(f, parentfiledata=parentfiledata)
 
     # merge
     for f, args, msg in actions.get(ACTION_MERGE, []):
         f1, f2, fa, move, anc = args
         if branchmerge:
             # We've done a branch merge, mark this file as merged
             # so that we properly record the merger later
             repo.dirstate.merge(f)
             if f1 != f2:  # copy/rename
                 if move:
                     repo.dirstate.remove(f1)
                 if f1 != f:
                     repo.dirstate.copy(f1, f)
                 else:
                     repo.dirstate.copy(f2, f)
         else:
             # We've update-merged a locally modified file, so
             # we set the dirstate to emulate a normal checkout
             # of that file some time in the past. Thus our
             # merge will appear as a normal local file
             # modification.
             if f2 == f:  # file not locally copied/moved
                 repo.dirstate.normallookup(f)
             if move:
                 repo.dirstate.drop(f1)
 
     # directory rename, move local
     for f, args, msg in actions.get(ACTION_DIR_RENAME_MOVE_LOCAL, []):
         f0, flag = args
         if branchmerge:
             repo.dirstate.add(f)
             repo.dirstate.remove(f0)
             repo.dirstate.copy(f0, f)
         else:
             repo.dirstate.normal(f)
             repo.dirstate.drop(f0)
 
     # directory rename, get
     for f, args, msg in actions.get(ACTION_LOCAL_DIR_RENAME_GET, []):
         f0, flag = args
         if branchmerge:
             repo.dirstate.add(f)
             repo.dirstate.copy(f0, f)
         else:
             repo.dirstate.normal(f)

mercurial/shelve.py

@@ -1,1188 +1,1188 b''
 # shelve.py - save/restore working directory state
 #
 # Copyright 2013 Facebook, Inc.
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 """save and restore changes to the working directory
 
 The "hg shelve" command saves changes made to the working directory
 and reverts those changes, resetting the working directory to a clean
 state.
 
 Later on, the "hg unshelve" command restores the changes saved by "hg
 shelve". Changes can be restored even after updating to a different
 parent, in which case Mercurial's merge machinery will resolve any
 conflicts if necessary.
 
 You can have more than one shelved change outstanding at a time; each
 shelved change has a distinct name. For details, see the help for "hg
 shelve".
 """
 from __future__ import absolute_import
 
 import collections
 import errno
 import itertools
 import stat
 
 from .i18n import _
 from .node import (
     bin,
     hex,
     nullid,
     nullrev,
 )
 from . import (
     bookmarks,
     bundle2,
     changegroup,
     cmdutil,
     discovery,
     error,
     exchange,
     hg,
     lock as lockmod,
     mdiff,
     merge,
     mergestate as mergestatemod,
     patch,
     phases,
     pycompat,
     repair,
     scmutil,
     templatefilters,
     util,
     vfs as vfsmod,
 )
 from .utils import (
     dateutil,
     stringutil,
 )
 
 backupdir = b'shelve-backup'
 shelvedir = b'shelved'
 shelvefileextensions = [b'hg', b'patch', b'shelve']
 
 # we never need the user, so we use a
 # generic user for all shelve operations
 shelveuser = b'shelve@localhost'
 
 
 class ShelfDir(object):
     def __init__(self, repo, for_backups=False):
         if for_backups:
             self.vfs = vfsmod.vfs(repo.vfs.join(backupdir))
         else:
             self.vfs = vfsmod.vfs(repo.vfs.join(shelvedir))
 
     def get(self, name):
         return Shelf(self.vfs, name)
 
     def listshelves(self):
         """return all shelves in repo as list of (time, name)"""
         try:
             names = self.vfs.listdir()
         except OSError as err:
             if err.errno != errno.ENOENT:
                 raise
             return []
         info = []
         seen = set()
         for filename in names:
             name = filename.rsplit(b'.', 1)[0]
             if name in seen:
                 continue
             seen.add(name)
             shelf = self.get(name)
             if not shelf.exists():
                 continue
             mtime = shelf.mtime()
             info.append((mtime, name))
         return sorted(info, reverse=True)
 
 
 class Shelf(object):
     """Represents a shelf, including possibly multiple files storing it.
 
     Old shelves will have a .patch and a .hg file. Newer shelves will
     also have a .shelve file. This class abstracts away some of the
     differences and lets you work with the shelf as a whole.
     """
 
     def __init__(self, vfs, name):
         self.vfs = vfs
         self.name = name
 
     def exists(self):
         return self.vfs.exists(self.name + b'.patch') and self.vfs.exists(
             self.name + b'.hg'
         )
 
     def mtime(self):
         return self.vfs.stat(self.name + b'.patch')[stat.ST_MTIME]
 
     def writeinfo(self, info):
         scmutil.simplekeyvaluefile(self.vfs, self.name + b'.shelve').write(info)
 
     def hasinfo(self):
         return self.vfs.exists(self.name + b'.shelve')
 
     def readinfo(self):
         return scmutil.simplekeyvaluefile(
             self.vfs, self.name + b'.shelve'
         ).read()
 
     def writebundle(self, repo, bases, node):
         cgversion = changegroup.safeversion(repo)
         if cgversion == b'01':
             btype = b'HG10BZ'
             compression = None
         else:
             btype = b'HG20'
             compression = b'BZ'
 
         repo = repo.unfiltered()
 
         outgoing = discovery.outgoing(
             repo, missingroots=bases, ancestorsof=[node]
         )
         cg = changegroup.makechangegroup(repo, outgoing, cgversion, b'shelve')
 
         bundle_filename = self.vfs.join(self.name + b'.hg')
         bundle2.writebundle(
             repo.ui,
             cg,
             bundle_filename,
             btype,
             self.vfs,
             compression=compression,
         )
 
     def applybundle(self, repo, tr):
         filename = self.name + b'.hg'
         fp = self.vfs(filename)
         try:
             targetphase = phases.internal
             if not phases.supportinternal(repo):
                 targetphase = phases.secret
             gen = exchange.readbundle(repo.ui, fp, filename, self.vfs)
             pretip = repo[b'tip']
             bundle2.applybundle(
                 repo,
                 gen,
                 tr,
                 source=b'unshelve',
                 url=b'bundle:' + self.vfs.join(filename),
                 targetphase=targetphase,
             )
             shelvectx = repo[b'tip']
             if pretip == shelvectx:
                 shelverev = tr.changes[b'revduplicates'][-1]
                 shelvectx = repo[shelverev]
             return shelvectx
         finally:
             fp.close()
 
     def open_patch(self, mode=b'rb'):
         return self.vfs(self.name + b'.patch', mode)
 
     def _backupfilename(self, backupvfs, filename):
         def gennames(base):
             yield base
             base, ext = base.rsplit(b'.', 1)
             for i in itertools.count(1):
                 yield b'%s-%d.%s' % (base, i, ext)
 
         for n in gennames(filename):
             if not backupvfs.exists(n):
                 return backupvfs.join(n)
 
     def movetobackup(self, backupvfs):
         if not backupvfs.isdir():
             backupvfs.makedir()
         for suffix in shelvefileextensions:
             filename = self.name + b'.' + suffix
             if self.vfs.exists(filename):
                 util.rename(
                     self.vfs.join(filename),
                     self._backupfilename(backupvfs, filename),
                 )
 
     def delete(self):
         for ext in shelvefileextensions:
             self.vfs.tryunlink(self.name + b'.' + ext)
 
 
 class shelvedstate(object):
     """Handle persistence during unshelving operations.
 
     Handles saving and restoring a shelved state. Ensures that different
     versions of a shelved state are possible and handles them appropriately.
     """
 
     _version = 2
     _filename = b'shelvedstate'
     _keep = b'keep'
     _nokeep = b'nokeep'
     # colon is essential to differentiate from a real bookmark name
     _noactivebook = b':no-active-bookmark'
     _interactive = b'interactive'
 
     @classmethod
     def _verifyandtransform(cls, d):
         """Some basic shelvestate syntactic verification and transformation"""
         try:
             d[b'originalwctx'] = bin(d[b'originalwctx'])
             d[b'pendingctx'] = bin(d[b'pendingctx'])
             d[b'parents'] = [bin(h) for h in d[b'parents'].split(b' ')]
             d[b'nodestoremove'] = [
                 bin(h) for h in d[b'nodestoremove'].split(b' ')
             ]
         except (ValueError, TypeError, KeyError) as err:
             raise error.CorruptedState(stringutil.forcebytestr(err))
 
     @classmethod
     def _getversion(cls, repo):
         """Read version information from shelvestate file"""
         fp = repo.vfs(cls._filename)
         try:
             version = int(fp.readline().strip())
         except ValueError as err:
             raise error.CorruptedState(stringutil.forcebytestr(err))
         finally:
             fp.close()
         return version
 
     @classmethod
     def _readold(cls, repo):
         """Read the old position-based version of a shelvestate file"""
         # Order is important, because old shelvestate file uses it
         # to detemine values of fields (i.g. name is on the second line,
         # originalwctx is on the third and so forth). Please do not change.
         keys = [
             b'version',
             b'name',
             b'originalwctx',
             b'pendingctx',
             b'parents',
             b'nodestoremove',
             b'branchtorestore',
             b'keep',
             b'activebook',
         ]
         # this is executed only seldomly, so it is not a big deal
         # that we open this file twice
         fp = repo.vfs(cls._filename)
         d = {}
         try:
             for key in keys:
                 d[key] = fp.readline().strip()
         finally:
             fp.close()
         return d
 
     @classmethod
     def load(cls, repo):
         version = cls._getversion(repo)
         if version < cls._version:
             d = cls._readold(repo)
         elif version == cls._version:
             d = scmutil.simplekeyvaluefile(repo.vfs, cls._filename).read(
                 firstlinenonkeyval=True
             )
         else:
             raise error.Abort(
                 _(
                     b'this version of shelve is incompatible '
                     b'with the version used in this repo'
                 )
             )
 
         cls._verifyandtransform(d)
         try:
             obj = cls()
             obj.name = d[b'name']
             obj.wctx = repo[d[b'originalwctx']]
             obj.pendingctx = repo[d[b'pendingctx']]
             obj.parents = d[b'parents']
             obj.nodestoremove = d[b'nodestoremove']
             obj.branchtorestore = d.get(b'branchtorestore', b'')
             obj.keep = d.get(b'keep') == cls._keep
             obj.activebookmark = b''
             if d.get(b'activebook', b'') != cls._noactivebook:
                 obj.activebookmark = d.get(b'activebook', b'')
             obj.interactive = d.get(b'interactive') == cls._interactive
         except (error.RepoLookupError, KeyError) as err:
             raise error.CorruptedState(pycompat.bytestr(err))
 
         return obj
 
     @classmethod
     def save(
         cls,
         repo,
         name,
         originalwctx,
         pendingctx,
         nodestoremove,
         branchtorestore,
         keep=False,
         activebook=b'',
         interactive=False,
     ):
         info = {
             b"name": name,
             b"originalwctx": hex(originalwctx.node()),
             b"pendingctx": hex(pendingctx.node()),
             b"parents": b' '.join([hex(p) for p in repo.dirstate.parents()]),
             b"nodestoremove": b' '.join([hex(n) for n in nodestoremove]),
             b"branchtorestore": branchtorestore,
             b"keep": cls._keep if keep else cls._nokeep,
             b"activebook": activebook or cls._noactivebook,
         }
         if interactive:
             info[b'interactive'] = cls._interactive
         scmutil.simplekeyvaluefile(repo.vfs, cls._filename).write(
             info, firstline=(b"%d" % cls._version)
         )
 
     @classmethod
     def clear(cls, repo):
         repo.vfs.unlinkpath(cls._filename, ignoremissing=True)
 
 
 def cleanupoldbackups(repo):
     maxbackups = repo.ui.configint(b'shelve', b'maxbackups')
     backup_dir = ShelfDir(repo, for_backups=True)
     hgfiles = backup_dir.listshelves()
     if maxbackups > 0 and maxbackups < len(hgfiles):
         bordermtime = hgfiles[maxbackups - 1][0]
     else:
         bordermtime = None
     for mtime, name in hgfiles[maxbackups:]:
         if mtime == bordermtime:
             # keep it, because timestamp can't decide exact order of backups
             continue
         backup_dir.get(name).delete()
 
 
 def _backupactivebookmark(repo):
     activebookmark = repo._activebookmark
     if activebookmark:
         bookmarks.deactivate(repo)
     return activebookmark
 
 
 def _restoreactivebookmark(repo, mark):
     if mark:
         bookmarks.activate(repo, mark)
 
 
 def _aborttransaction(repo, tr):
     """Abort current transaction for shelve/unshelve, but keep dirstate"""
     dirstatebackupname = b'dirstate.shelve'
     repo.dirstate.savebackup(tr, dirstatebackupname)
     tr.abort()
     repo.dirstate.restorebackup(None, dirstatebackupname)
 
 
 def getshelvename(repo, parent, opts):
     """Decide on the name this shelve is going to have"""
 
     def gennames():
         yield label
         for i in itertools.count(1):
             yield b'%s-%02d' % (label, i)
 
     name = opts.get(b'name')
     label = repo._activebookmark or parent.branch() or b'default'
     # slashes aren't allowed in filenames, therefore we rename it
     label = label.replace(b'/', b'_')
     label = label.replace(b'\\', b'_')
     # filenames must not start with '.' as it should not be hidden
     if label.startswith(b'.'):
         label = label.replace(b'.', b'_', 1)
 
     if name:
         if ShelfDir(repo).get(name).exists():
             e = _(b"a shelved change named '%s' already exists") % name
             raise error.Abort(e)
 
         # ensure we are not creating a subdirectory or a hidden file
         if b'/' in name or b'\\' in name:
             raise error.Abort(
                 _(b'shelved change names can not contain slashes')
             )
         if name.startswith(b'.'):
             raise error.Abort(_(b"shelved change names can not start with '.'"))
 
     else:
         shelf_dir = ShelfDir(repo)
         for n in gennames():
             if not shelf_dir.get(n).exists():
                 name = n
                 break
 
     return name
 
 
 def mutableancestors(ctx):
     """return all mutable ancestors for ctx (included)
 
     Much faster than the revset ancestors(ctx) & draft()"""
     seen = {nullrev}
     visit = collections.deque()
     visit.append(ctx)
     while visit:
         ctx = visit.popleft()
         yield ctx.node()
         for parent in ctx.parents():
             rev = parent.rev()
             if rev not in seen:
                 seen.add(rev)
                 if parent.mutable():
                     visit.append(parent)
 
 
 def getcommitfunc(extra, interactive, editor=False):
     def commitfunc(ui, repo, message, match, opts):
         hasmq = util.safehasattr(repo, b'mq')
         if hasmq:
             saved, repo.mq.checkapplied = repo.mq.checkapplied, False
 
         targetphase = phases.internal
         if not phases.supportinternal(repo):
             targetphase = phases.secret
         overrides = {(b'phases', b'new-commit'): targetphase}
         try:
             editor_ = False
             if editor:
                 editor_ = cmdutil.getcommiteditor(
                     editform=b'shelve.shelve', **pycompat.strkwargs(opts)
                 )
             with repo.ui.configoverride(overrides):
                 return repo.commit(
                     message,
                     shelveuser,
                     opts.get(b'date'),
                     match,
                     editor=editor_,
                     extra=extra,
                 )
         finally:
             if hasmq:
                 repo.mq.checkapplied = saved
 
     def interactivecommitfunc(ui, repo, *pats, **opts):
         opts = pycompat.byteskwargs(opts)
         match = scmutil.match(repo[b'.'], pats, {})
         message = opts[b'message']
         return commitfunc(ui, repo, message, match, opts)
 
     return interactivecommitfunc if interactive else commitfunc
 
 
 def _nothingtoshelvemessaging(ui, repo, pats, opts):
     stat = repo.status(match=scmutil.match(repo[None], pats, opts))
     if stat.deleted:
         ui.status(
             _(b"nothing changed (%d missing files, see 'hg status')\n")
             % len(stat.deleted)
         )
     else:
         ui.status(_(b"nothing changed\n"))
 
 
 def _shelvecreatedcommit(repo, node, name, match):
     info = {b'node': hex(node)}
     shelf = ShelfDir(repo).get(name)
     shelf.writeinfo(info)
     bases = list(mutableancestors(repo[node]))
     shelf.writebundle(repo, bases, node)
     with shelf.open_patch(b'wb') as fp:
         cmdutil.exportfile(
             repo, [node], fp, opts=mdiff.diffopts(git=True), match=match
         )
 
 
 def _includeunknownfiles(repo, pats, opts, extra):
     s = repo.status(match=scmutil.match(repo[None], pats, opts), unknown=True)
     if s.unknown:
         extra[b'shelve_unknown'] = b'\0'.join(s.unknown)
         repo[None].add(s.unknown)
 
 
 def _finishshelve(repo, tr):
     if phases.supportinternal(repo):
         tr.close()
     else:
         _aborttransaction(repo, tr)
 
 
 def createcmd(ui, repo, pats, opts):
     """subcommand that creates a new shelve"""
     with repo.wlock():
         cmdutil.checkunfinished(repo)
         return _docreatecmd(ui, repo, pats, opts)
 
 
 def _docreatecmd(ui, repo, pats, opts):
     wctx = repo[None]
     parents = wctx.parents()
     parent = parents[0]
     origbranch = wctx.branch()
 
-    if parent.node() != nullid:
+    if parent.rev() != nullrev:
         desc = b"changes to: %s" % parent.description().split(b'\n', 1)[0]
     else:
         desc = b'(changes in empty repository)'
 
     if not opts.get(b'message'):
         opts[b'message'] = desc
 
     lock = tr = activebookmark = None
     try:
         lock = repo.lock()
 
         # use an uncommitted transaction to generate the bundle to avoid
         # pull races. ensure we don't print the abort message to stderr.
         tr = repo.transaction(b'shelve', report=lambda x: None)
 
         interactive = opts.get(b'interactive', False)
         includeunknown = opts.get(b'unknown', False) and not opts.get(
             b'addremove', False
         )
 
         name = getshelvename(repo, parent, opts)
         activebookmark = _backupactivebookmark(repo)
         extra = {b'internal': b'shelve'}
         if includeunknown:
             _includeunknownfiles(repo, pats, opts, extra)
 
         if _iswctxonnewbranch(repo) and not _isbareshelve(pats, opts):
             # In non-bare shelve we don't store newly created branch
             # at bundled commit
             repo.dirstate.setbranch(repo[b'.'].branch())
 
         commitfunc = getcommitfunc(extra, interactive, editor=True)
         if not interactive:
             node = cmdutil.commit(ui, repo, commitfunc, pats, opts)
         else:
             node = cmdutil.dorecord(
                 ui,
                 repo,
                 commitfunc,
                 None,
                 False,
                 cmdutil.recordfilter,
                 *pats,
                 **pycompat.strkwargs(opts)
             )
         if not node:
             _nothingtoshelvemessaging(ui, repo, pats, opts)
             return 1
 
         # Create a matcher so that prefetch doesn't attempt to fetch
         # the entire repository pointlessly, and as an optimisation
         # for movedirstate, if needed.
         match = scmutil.matchfiles(repo, repo[node].files())
         _shelvecreatedcommit(repo, node, name, match)
 
         ui.status(_(b'shelved as %s\n') % name)
         if opts[b'keep']:
             with repo.dirstate.parentchange():
                 scmutil.movedirstate(repo, parent, match)
         else:
             hg.update(repo, parent.node())
             ms = mergestatemod.mergestate.read(repo)
             if not ms.unresolvedcount():
                 ms.reset()
 
         if origbranch != repo[b'.'].branch() and not _isbareshelve(pats, opts):
             repo.dirstate.setbranch(origbranch)
 
         _finishshelve(repo, tr)
     finally:
         _restoreactivebookmark(repo, activebookmark)
         lockmod.release(tr, lock)
 
 
 def _isbareshelve(pats, opts):
     return (
         not pats
         and not opts.get(b'interactive', False)
         and not opts.get(b'include', False)
         and not opts.get(b'exclude', False)
     )
 
 
 def _iswctxonnewbranch(repo):
     return repo[None].branch() != repo[b'.'].branch()
 
 
 def cleanupcmd(ui, repo):
     """subcommand that deletes all shelves"""
 
     with repo.wlock():
         shelf_dir = ShelfDir(repo)
         backupvfs = vfsmod.vfs(repo.vfs.join(backupdir))
         for _mtime, name in shelf_dir.listshelves():
             shelf_dir.get(name).movetobackup(backupvfs)
             cleanupoldbackups(repo)
 
 
 def deletecmd(ui, repo, pats):
     """subcommand that deletes a specific shelve"""
     if not pats:
         raise error.InputError(_(b'no shelved changes specified!'))
     with repo.wlock():
         backupvfs = vfsmod.vfs(repo.vfs.join(backupdir))
         for name in pats:
             shelf = ShelfDir(repo).get(name)
             if not shelf.exists():
                 raise error.InputError(
                     _(b"shelved change '%s' not found") % name
                 )
             shelf.movetobackup(backupvfs)
             cleanupoldbackups(repo)
 
 
 def listcmd(ui, repo, pats, opts):
     """subcommand that displays the list of shelves"""
     pats = set(pats)
     width = 80
     if not ui.plain():
         width = ui.termwidth()
     namelabel = b'shelve.newest'
     ui.pager(b'shelve')
     shelf_dir = ShelfDir(repo)
     for mtime, name in shelf_dir.listshelves():
         if pats and name not in pats:
             continue
         ui.write(name, label=namelabel)
         namelabel = b'shelve.name'
         if ui.quiet:
             ui.write(b'\n')
             continue
         ui.write(b' ' * (16 - len(name)))
         used = 16
         date = dateutil.makedate(mtime)
         age = b'(%s)' % templatefilters.age(date, abbrev=True)
         ui.write(age, label=b'shelve.age')
         ui.write(b' ' * (12 - len(age)))
         used += 12
         with shelf_dir.get(name).open_patch() as fp:
             while True:
                 line = fp.readline()
                 if not line:
                     break
                 if not line.startswith(b'#'):
                     desc = line.rstrip()
                     if ui.formatted():
                         desc = stringutil.ellipsis(desc, width - used)
                     ui.write(desc)
                     break
             ui.write(b'\n')
             if not (opts[b'patch'] or opts[b'stat']):
                 continue
             difflines = fp.readlines()
             if opts[b'patch']:
                 for chunk, label in patch.difflabel(iter, difflines):
                     ui.write(chunk, label=label)
             if opts[b'stat']:
                 for chunk, label in patch.diffstatui(difflines, width=width):
                     ui.write(chunk, label=label)
 
 
 def patchcmds(ui, repo, pats, opts):
     """subcommand that displays shelves"""
     shelf_dir = ShelfDir(repo)
     if len(pats) == 0:
         shelves = shelf_dir.listshelves()
         if not shelves:
             raise error.Abort(_(b"there are no shelves to show"))
         mtime, name = shelves[0]
         pats = [name]
 
     for shelfname in pats:
         if not shelf_dir.get(shelfname).exists():
             raise error.Abort(_(b"cannot find shelf %s") % shelfname)
 
     listcmd(ui, repo, pats, opts)
 
 
 def checkparents(repo, state):
     """check parent while resuming an unshelve"""
     if state.parents != repo.dirstate.parents():
         raise error.Abort(
             _(b'working directory parents do not match unshelve state')
         )
 
 
 def _loadshelvedstate(ui, repo, opts):
     try:
         state = shelvedstate.load(repo)
         if opts.get(b'keep') is None:
             opts[b'keep'] = state.keep
     except IOError as err:
         if err.errno != errno.ENOENT:
             raise
         cmdutil.wrongtooltocontinue(repo, _(b'unshelve'))
     except error.CorruptedState as err:
         ui.debug(pycompat.bytestr(err) + b'\n')
         if opts.get(b'continue'):
             msg = _(b'corrupted shelved state file')
             hint = _(
                 b'please run hg unshelve --abort to abort unshelve '
                 b'operation'
             )
             raise error.Abort(msg, hint=hint)
         elif opts.get(b'abort'):
             shelvedstate.clear(repo)
             raise error.Abort(
                 _(
                     b'could not read shelved state file, your '
                     b'working copy may be in an unexpected state\n'
                     b'please update to some commit\n'
                 )
             )
     return state
 
 
 def unshelveabort(ui, repo, state):
     """subcommand that abort an in-progress unshelve"""
     with repo.lock():
         try:
             checkparents(repo, state)
 
             merge.clean_update(state.pendingctx)
             if state.activebookmark and state.activebookmark in repo._bookmarks:
                 bookmarks.activate(repo, state.activebookmark)
             mergefiles(ui, repo, state.wctx, state.pendingctx)
             if not phases.supportinternal(repo):
                 repair.strip(
                     ui, repo, state.nodestoremove, backup=False, topic=b'shelve'
                 )
         finally:
             shelvedstate.clear(repo)
             ui.warn(_(b"unshelve of '%s' aborted\n") % state.name)
 
 
 def hgabortunshelve(ui, repo):
     """logic to  abort unshelve using 'hg abort"""
     with repo.wlock():
         state = _loadshelvedstate(ui, repo, {b'abort': True})
         return unshelveabort(ui, repo, state)
 
 
 def mergefiles(ui, repo, wctx, shelvectx):
     """updates to wctx and merges the changes from shelvectx into the
     dirstate."""
     with ui.configoverride({(b'ui', b'quiet'): True}):
         hg.update(repo, wctx.node())
         ui.pushbuffer(True)
         cmdutil.revert(ui, repo, shelvectx)
         ui.popbuffer()
 
 
 def restorebranch(ui, repo, branchtorestore):
     if branchtorestore and branchtorestore != repo.dirstate.branch():
         repo.dirstate.setbranch(branchtorestore)
         ui.status(
             _(b'marked working directory as branch %s\n') % branchtorestore
         )
 
 
 def unshelvecleanup(ui, repo, name, opts):
     """remove related files after an unshelve"""
     if not opts.get(b'keep'):
         backupvfs = vfsmod.vfs(repo.vfs.join(backupdir))
         ShelfDir(repo).get(name).movetobackup(backupvfs)
         cleanupoldbackups(repo)
 
 
 def unshelvecontinue(ui, repo, state, opts):
     """subcommand to continue an in-progress unshelve"""
     # We're finishing off a merge. First parent is our original
     # parent, second is the temporary "fake" commit we're unshelving.
     interactive = state.interactive
     basename = state.name
     with repo.lock():
         checkparents(repo, state)
         ms = mergestatemod.mergestate.read(repo)
         if ms.unresolvedcount():
             raise error.Abort(
                 _(b"unresolved conflicts, can't continue"),
                 hint=_(b"see 'hg resolve', then 'hg unshelve --continue'"),
             )
 
         shelvectx = repo[state.parents[1]]
         pendingctx = state.pendingctx
 
         with repo.dirstate.parentchange():
             repo.setparents(state.pendingctx.node(), nullid)
             repo.dirstate.write(repo.currenttransaction())
 
         targetphase = phases.internal
         if not phases.supportinternal(repo):
             targetphase = phases.secret
         overrides = {(b'phases', b'new-commit'): targetphase}
         with repo.ui.configoverride(overrides, b'unshelve'):
             with repo.dirstate.parentchange():
                 repo.setparents(state.parents[0], nullid)
                 newnode, ispartialunshelve = _createunshelvectx(
                     ui, repo, shelvectx, basename, interactive, opts
                 )
 
         if newnode is None:
             shelvectx = state.pendingctx
             msg = _(
                 b'note: unshelved changes already existed '
                 b'in the working copy\n'
             )
             ui.status(msg)
         else:
             # only strip the shelvectx if we produced one
             state.nodestoremove.append(newnode)
             shelvectx = repo[newnode]
 
         merge.update(pendingctx)
         mergefiles(ui, repo, state.wctx, shelvectx)
         restorebranch(ui, repo, state.branchtorestore)
 
         if not phases.supportinternal(repo):
             repair.strip(
                 ui, repo, state.nodestoremove, backup=False, topic=b'shelve'
             )
         shelvedstate.clear(repo)
         if not ispartialunshelve:
             unshelvecleanup(ui, repo, state.name, opts)
         _restoreactivebookmark(repo, state.activebookmark)
         ui.status(_(b"unshelve of '%s' complete\n") % state.name)
 
 
 def hgcontinueunshelve(ui, repo):
     """logic to resume unshelve using 'hg continue'"""
     with repo.wlock():
         state = _loadshelvedstate(ui, repo, {b'continue': True})
         return unshelvecontinue(ui, repo, state, {b'keep': state.keep})
 
 
 def _commitworkingcopychanges(ui, repo, opts, tmpwctx):
     """Temporarily commit working copy changes before moving unshelve commit"""
     # Store pending changes in a commit and remember added in case a shelve
     # contains unknown files that are part of the pending change
     s = repo.status()
     addedbefore = frozenset(s.added)
     if not (s.modified or s.added or s.removed):
         return tmpwctx, addedbefore
     ui.status(
         _(
             b"temporarily committing pending changes "
             b"(restore with 'hg unshelve --abort')\n"
         )
     )
     extra = {b'internal': b'shelve'}
     commitfunc = getcommitfunc(extra=extra, interactive=False, editor=False)
     tempopts = {}
     tempopts[b'message'] = b"pending changes temporary commit"
     tempopts[b'date'] = opts.get(b'date')
     with ui.configoverride({(b'ui', b'quiet'): True}):
         node = cmdutil.commit(ui, repo, commitfunc, [], tempopts)
     tmpwctx = repo[node]
     return tmpwctx, addedbefore
 
 
 def _unshelverestorecommit(ui, repo, tr, basename):
     """Recreate commit in the repository during the unshelve"""
     repo = repo.unfiltered()
     node = None
     shelf = ShelfDir(repo).get(basename)
     if shelf.hasinfo():
         node = shelf.readinfo()[b'node']
     if node is None or node not in repo:
         with ui.configoverride({(b'ui', b'quiet'): True}):
             shelvectx = shelf.applybundle(repo, tr)
         # We might not strip the unbundled changeset, so we should keep track of
         # the unshelve node in case we need to reuse it (eg: unshelve --keep)
         if node is None:
             info = {b'node': hex(shelvectx.node())}
             shelf.writeinfo(info)
     else:
         shelvectx = repo[node]
 
     return repo, shelvectx
 
 
 def _createunshelvectx(ui, repo, shelvectx, basename, interactive, opts):
     """Handles the creation of unshelve commit and updates the shelve if it
     was partially unshelved.
 
     If interactive is:
 
       * False: Commits all the changes in the working directory.
       * True: Prompts the user to select changes to unshelve and commit them.
               Update the shelve with remaining changes.
 
     Returns the node of the new commit formed and a bool indicating whether
     the shelve was partially unshelved.Creates a commit ctx to unshelve
     interactively or non-interactively.
 
     The user might want to unshelve certain changes only from the stored
     shelve in interactive. So, we would create two commits. One with requested
     changes to unshelve at that time and the latter is shelved for future.
 
     Here, we return both the newnode which is created interactively and a
     bool to know whether the shelve is partly done or completely done.
     """
     opts[b'message'] = shelvectx.description()
     opts[b'interactive-unshelve'] = True
     pats = []
     if not interactive:
         newnode = repo.commit(
             text=shelvectx.description(),
             extra=shelvectx.extra(),
             user=shelvectx.user(),
             date=shelvectx.date(),
         )
         return newnode, False
 
     commitfunc = getcommitfunc(shelvectx.extra(), interactive=True, editor=True)
     newnode = cmdutil.dorecord(
         ui,
         repo,
         commitfunc,
         None,
         False,
         cmdutil.recordfilter,
         *pats,
         **pycompat.strkwargs(opts)
     )
     snode = repo.commit(
         text=shelvectx.description(),
         extra=shelvectx.extra(),
         user=shelvectx.user(),
     )
     if snode:
         m = scmutil.matchfiles(repo, repo[snode].files())
         _shelvecreatedcommit(repo, snode, basename, m)
 
     return newnode, bool(snode)
 
 
 def _rebaserestoredcommit(
     ui,
     repo,
     opts,
     tr,
     oldtiprev,
     basename,
     pctx,
     tmpwctx,
     shelvectx,
     branchtorestore,
     activebookmark,
 ):
     """Rebase restored commit from its original location to a destination"""
     # If the shelve is not immediately on top of the commit
     # we'll be merging with, rebase it to be on top.
     interactive = opts.get(b'interactive')
     if tmpwctx.node() == shelvectx.p1().node() and not interactive:
         # We won't skip on interactive mode because, the user might want to
         # unshelve certain changes only.
         return shelvectx, False
 
     overrides = {
         (b'ui', b'forcemerge'): opts.get(b'tool', b''),
         (b'phases', b'new-commit'): phases.secret,
     }
     with repo.ui.configoverride(overrides, b'unshelve'):
         ui.status(_(b'rebasing shelved changes\n'))
         stats = merge.graft(
             repo,
             shelvectx,
             labels=[b'working-copy', b'shelve'],
             keepconflictparent=True,
         )
         if stats.unresolvedcount:
             tr.close()
 
             nodestoremove = [
                 repo.changelog.node(rev)
                 for rev in pycompat.xrange(oldtiprev, len(repo))
             ]
             shelvedstate.save(
                 repo,
                 basename,
                 pctx,
                 tmpwctx,
                 nodestoremove,
                 branchtorestore,
                 opts.get(b'keep'),
                 activebookmark,
                 interactive,
             )
             raise error.ConflictResolutionRequired(b'unshelve')
 
         with repo.dirstate.parentchange():
             repo.setparents(tmpwctx.node(), nullid)
             newnode, ispartialunshelve = _createunshelvectx(
                 ui, repo, shelvectx, basename, interactive, opts
             )
 
         if newnode is None:
             shelvectx = tmpwctx
             msg = _(
                 b'note: unshelved changes already existed '
                 b'in the working copy\n'
             )
             ui.status(msg)
         else:
             shelvectx = repo[newnode]
             merge.update(tmpwctx)
 
     return shelvectx, ispartialunshelve
 
 
 def _forgetunknownfiles(repo, shelvectx, addedbefore):
     # Forget any files that were unknown before the shelve, unknown before
     # unshelve started, but are now added.
     shelveunknown = shelvectx.extra().get(b'shelve_unknown')
     if not shelveunknown:
         return
     shelveunknown = frozenset(shelveunknown.split(b'\0'))
     addedafter = frozenset(repo.status().added)
     toforget = (addedafter & shelveunknown) - addedbefore
     repo[None].forget(toforget)
 
 
 def _finishunshelve(repo, oldtiprev, tr, activebookmark):
     _restoreactivebookmark(repo, activebookmark)
     # The transaction aborting will strip all the commits for us,
     # but it doesn't update the inmemory structures, so addchangegroup
     # hooks still fire and try to operate on the missing commits.
     # Clean up manually to prevent this.
     repo.unfiltered().changelog.strip(oldtiprev, tr)
     _aborttransaction(repo, tr)
 
 
 def _checkunshelveuntrackedproblems(ui, repo, shelvectx):
     """Check potential problems which may result from working
     copy having untracked changes."""
     wcdeleted = set(repo.status().deleted)
     shelvetouched = set(shelvectx.files())
     intersection = wcdeleted.intersection(shelvetouched)
     if intersection:
         m = _(b"shelved change touches missing files")
         hint = _(b"run hg status to see which files are missing")
         raise error.Abort(m, hint=hint)
 
 
 def unshelvecmd(ui, repo, *shelved, **opts):
     opts = pycompat.byteskwargs(opts)
     abortf = opts.get(b'abort')
     continuef = opts.get(b'continue')
     interactive = opts.get(b'interactive')
     if not abortf and not continuef:
         cmdutil.checkunfinished(repo)
     shelved = list(shelved)
     if opts.get(b"name"):
         shelved.append(opts[b"name"])
 
     if interactive and opts.get(b'keep'):
         raise error.InputError(
             _(b'--keep on --interactive is not yet supported')
         )
     if abortf or continuef:
         if abortf and continuef:
             raise error.InputError(_(b'cannot use both abort and continue'))
         if shelved:
             raise error.InputError(
                 _(
                     b'cannot combine abort/continue with '
                     b'naming a shelved change'
                 )
             )
         if abortf and opts.get(b'tool', False):
             ui.warn(_(b'tool option will be ignored\n'))
 
         state = _loadshelvedstate(ui, repo, opts)
         if abortf:
             return unshelveabort(ui, repo, state)
         elif continuef and interactive:
             raise error.InputError(
                 _(b'cannot use both continue and interactive')
             )
         elif continuef:
             return unshelvecontinue(ui, repo, state, opts)
     elif len(shelved) > 1:
         raise error.InputError(_(b'can only unshelve one change at a time'))
     elif not shelved:
         shelved = ShelfDir(repo).listshelves()
         if not shelved:
             raise error.StateError(_(b'no shelved changes to apply!'))
         basename = shelved[0][1]
         ui.status(_(b"unshelving change '%s'\n") % basename)
     else:
         basename = shelved[0]
 
     if not ShelfDir(repo).get(basename).exists():
         raise error.InputError(_(b"shelved change '%s' not found") % basename)
 
     return _dounshelve(ui, repo, basename, opts)
 
 
 def _dounshelve(ui, repo, basename, opts):
     repo = repo.unfiltered()
     lock = tr = None
     try:
         lock = repo.lock()
         tr = repo.transaction(b'unshelve', report=lambda x: None)
         oldtiprev = len(repo)
 
         pctx = repo[b'.']
         tmpwctx = pctx
         # The goal is to have a commit structure like so:
         # ...-> pctx -> tmpwctx -> shelvectx
         # where tmpwctx is an optional commit with the user's pending changes
         # and shelvectx is the unshelved changes. Then we merge it all down
         # to the original pctx.
 
         activebookmark = _backupactivebookmark(repo)
         tmpwctx, addedbefore = _commitworkingcopychanges(
             ui, repo, opts, tmpwctx
         )
         repo, shelvectx = _unshelverestorecommit(ui, repo, tr, basename)
         _checkunshelveuntrackedproblems(ui, repo, shelvectx)
         branchtorestore = b''
         if shelvectx.branch() != shelvectx.p1().branch():
             branchtorestore = shelvectx.branch()
 
         shelvectx, ispartialunshelve = _rebaserestoredcommit(
             ui,
             repo,
             opts,
             tr,
             oldtiprev,
             basename,
             pctx,
             tmpwctx,
             shelvectx,
             branchtorestore,
             activebookmark,
         )
         overrides = {(b'ui', b'forcemerge'): opts.get(b'tool', b'')}
         with ui.configoverride(overrides, b'unshelve'):
             mergefiles(ui, repo, pctx, shelvectx)
         restorebranch(ui, repo, branchtorestore)
         shelvedstate.clear(repo)
         _finishunshelve(repo, oldtiprev, tr, activebookmark)
         _forgetunknownfiles(repo, shelvectx, addedbefore)
         if not ispartialunshelve:
             unshelvecleanup(ui, repo, basename, opts)
     finally:
         if tr:
             tr.release()
         lockmod.release(lock)

mercurial/simplemerge.py

@@ -1,566 +1,566 b''
 # Copyright (C) 2004, 2005 Canonical Ltd
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
 # the Free Software Foundation; either version 2 of the License, or
 # (at your option) any later version.
 #
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
 #
 # You should have received a copy of the GNU General Public License
 # along with this program; if not, see <http://www.gnu.org/licenses/>.
 
 # mbp: "you know that thing where cvs gives you conflict markers?"
 # s: "i hate that."
 
 from __future__ import absolute_import
 
 from .i18n import _
-from .node import nullid
+from .node import nullrev
 from . import (
     error,
     mdiff,
     pycompat,
     util,
 )
 from .utils import stringutil
 
 
 class CantReprocessAndShowBase(Exception):
     pass
 
 
 def intersect(ra, rb):
     """Given two ranges return the range where they intersect or None.
 
     >>> intersect((0, 10), (0, 6))
     (0, 6)
     >>> intersect((0, 10), (5, 15))
     (5, 10)
     >>> intersect((0, 10), (10, 15))
     >>> intersect((0, 9), (10, 15))
     >>> intersect((0, 9), (7, 15))
     (7, 9)
     """
     assert ra[0] <= ra[1]
     assert rb[0] <= rb[1]
 
     sa = max(ra[0], rb[0])
     sb = min(ra[1], rb[1])
     if sa < sb:
         return sa, sb
     else:
         return None
 
 
 def compare_range(a, astart, aend, b, bstart, bend):
     """Compare a[astart:aend] == b[bstart:bend], without slicing."""
     if (aend - astart) != (bend - bstart):
         return False
     for ia, ib in zip(
         pycompat.xrange(astart, aend), pycompat.xrange(bstart, bend)
     ):
         if a[ia] != b[ib]:
             return False
     else:
         return True
 
 
 class Merge3Text(object):
     """3-way merge of texts.
 
     Given strings BASE, OTHER, THIS, tries to produce a combined text
     incorporating the changes from both BASE->OTHER and BASE->THIS."""
 
     def __init__(self, basetext, atext, btext, base=None, a=None, b=None):
         self.basetext = basetext
         self.atext = atext
         self.btext = btext
         if base is None:
             base = mdiff.splitnewlines(basetext)
         if a is None:
             a = mdiff.splitnewlines(atext)
         if b is None:
             b = mdiff.splitnewlines(btext)
         self.base = base
         self.a = a
         self.b = b
 
     def merge_lines(
         self,
         name_a=None,
         name_b=None,
         name_base=None,
         start_marker=b'<<<<<<<',
         mid_marker=b'=======',
         end_marker=b'>>>>>>>',
         base_marker=None,
         localorother=None,
         minimize=False,
     ):
         """Return merge in cvs-like form."""
         self.conflicts = False
         newline = b'\n'
         if len(self.a) > 0:
             if self.a[0].endswith(b'\r\n'):
                 newline = b'\r\n'
             elif self.a[0].endswith(b'\r'):
                 newline = b'\r'
         if name_a and start_marker:
             start_marker = start_marker + b' ' + name_a
         if name_b and end_marker:
             end_marker = end_marker + b' ' + name_b
         if name_base and base_marker:
             base_marker = base_marker + b' ' + name_base
         merge_regions = self.merge_regions()
         if minimize:
             merge_regions = self.minimize(merge_regions)
         for t in merge_regions:
             what = t[0]
             if what == b'unchanged':
                 for i in range(t[1], t[2]):
                     yield self.base[i]
             elif what == b'a' or what == b'same':
                 for i in range(t[1], t[2]):
                     yield self.a[i]
             elif what == b'b':
                 for i in range(t[1], t[2]):
                     yield self.b[i]
             elif what == b'conflict':
                 if localorother == b'local':
                     for i in range(t[3], t[4]):
                         yield self.a[i]
                 elif localorother == b'other':
                     for i in range(t[5], t[6]):
                         yield self.b[i]
                 else:
                     self.conflicts = True
                     if start_marker is not None:
                         yield start_marker + newline
                     for i in range(t[3], t[4]):
                         yield self.a[i]
                     if base_marker is not None:
                         yield base_marker + newline
                         for i in range(t[1], t[2]):
                             yield self.base[i]
                     if mid_marker is not None:
                         yield mid_marker + newline
                     for i in range(t[5], t[6]):
                         yield self.b[i]
                     if end_marker is not None:
                         yield end_marker + newline
             else:
                 raise ValueError(what)
 
     def merge_groups(self):
         """Yield sequence of line groups.  Each one is a tuple:
 
         'unchanged', lines
              Lines unchanged from base
 
         'a', lines
              Lines taken from a
 
         'same', lines
              Lines taken from a (and equal to b)
 
         'b', lines
              Lines taken from b
 
         'conflict', base_lines, a_lines, b_lines
              Lines from base were changed to either a or b and conflict.
         """
         for t in self.merge_regions():
             what = t[0]
             if what == b'unchanged':
                 yield what, self.base[t[1] : t[2]]
             elif what == b'a' or what == b'same':
                 yield what, self.a[t[1] : t[2]]
             elif what == b'b':
                 yield what, self.b[t[1] : t[2]]
             elif what == b'conflict':
                 yield (
                     what,
                     self.base[t[1] : t[2]],
                     self.a[t[3] : t[4]],
                     self.b[t[5] : t[6]],
                 )
             else:
                 raise ValueError(what)
 
     def merge_regions(self):
         """Return sequences of matching and conflicting regions.
 
         This returns tuples, where the first value says what kind we
         have:
 
         'unchanged', start, end
              Take a region of base[start:end]
 
         'same', astart, aend
              b and a are different from base but give the same result
 
         'a', start, end
              Non-clashing insertion from a[start:end]
 
         'conflict', zstart, zend, astart, aend, bstart, bend
             Conflict between a and b, with z as common ancestor
 
         Method is as follows:
 
         The two sequences align only on regions which match the base
         and both descendants.  These are found by doing a two-way diff
         of each one against the base, and then finding the
         intersections between those regions.  These "sync regions"
         are by definition unchanged in both and easily dealt with.
 
         The regions in between can be in any of three cases:
         conflicted, or changed on only one side.
         """
 
         # section a[0:ia] has been disposed of, etc
         iz = ia = ib = 0
 
         for region in self.find_sync_regions():
             zmatch, zend, amatch, aend, bmatch, bend = region
             # print 'match base [%d:%d]' % (zmatch, zend)
 
             matchlen = zend - zmatch
             assert matchlen >= 0
             assert matchlen == (aend - amatch)
             assert matchlen == (bend - bmatch)
 
             len_a = amatch - ia
             len_b = bmatch - ib
             len_base = zmatch - iz
             assert len_a >= 0
             assert len_b >= 0
             assert len_base >= 0
 
             # print 'unmatched a=%d, b=%d' % (len_a, len_b)
 
             if len_a or len_b:
                 # try to avoid actually slicing the lists
                 equal_a = compare_range(
                     self.a, ia, amatch, self.base, iz, zmatch
                 )
                 equal_b = compare_range(
                     self.b, ib, bmatch, self.base, iz, zmatch
                 )
                 same = compare_range(self.a, ia, amatch, self.b, ib, bmatch)
 
                 if same:
                     yield b'same', ia, amatch
                 elif equal_a and not equal_b:
                     yield b'b', ib, bmatch
                 elif equal_b and not equal_a:
                     yield b'a', ia, amatch
                 elif not equal_a and not equal_b:
                     yield b'conflict', iz, zmatch, ia, amatch, ib, bmatch
                 else:
                     raise AssertionError(b"can't handle a=b=base but unmatched")
 
                 ia = amatch
                 ib = bmatch
             iz = zmatch
 
             # if the same part of the base was deleted on both sides
             # that's OK, we can just skip it.
 
             if matchlen > 0:
                 assert ia == amatch
                 assert ib == bmatch
                 assert iz == zmatch
 
                 yield b'unchanged', zmatch, zend
                 iz = zend
                 ia = aend
                 ib = bend
 
     def minimize(self, merge_regions):
         """Trim conflict regions of lines where A and B sides match.
 
         Lines where both A and B have made the same changes at the beginning
         or the end of each merge region are eliminated from the conflict
         region and are instead considered the same.
         """
         for region in merge_regions:
             if region[0] != b"conflict":
                 yield region
                 continue
             # pytype thinks this tuple contains only 3 things, but
             # that's clearly not true because this code successfully
             # executes. It might be wise to rework merge_regions to be
             # some kind of attrs type.
             (
                 issue,
                 z1,
                 z2,
                 a1,
                 a2,
                 b1,
                 b2,
             ) = region  # pytype: disable=bad-unpacking
             alen = a2 - a1
             blen = b2 - b1
 
             # find matches at the front
             ii = 0
             while (
                 ii < alen and ii < blen and self.a[a1 + ii] == self.b[b1 + ii]
             ):
                 ii += 1
             startmatches = ii
 
             # find matches at the end
             ii = 0
             while (
                 ii < alen
                 and ii < blen
                 and self.a[a2 - ii - 1] == self.b[b2 - ii - 1]
             ):
                 ii += 1
             endmatches = ii
 
             if startmatches > 0:
                 yield b'same', a1, a1 + startmatches
 
             yield (
                 b'conflict',
                 z1,
                 z2,
                 a1 + startmatches,
                 a2 - endmatches,
                 b1 + startmatches,
                 b2 - endmatches,
             )
 
             if endmatches > 0:
                 yield b'same', a2 - endmatches, a2
 
     def find_sync_regions(self):
         """Return a list of sync regions, where both descendants match the base.
 
         Generates a list of (base1, base2, a1, a2, b1, b2).  There is
         always a zero-length sync region at the end of all the files.
         """
 
         ia = ib = 0
         amatches = mdiff.get_matching_blocks(self.basetext, self.atext)
         bmatches = mdiff.get_matching_blocks(self.basetext, self.btext)
         len_a = len(amatches)
         len_b = len(bmatches)
 
         sl = []
 
         while ia < len_a and ib < len_b:
             abase, amatch, alen = amatches[ia]
             bbase, bmatch, blen = bmatches[ib]
 
             # there is an unconflicted block at i; how long does it
             # extend?  until whichever one ends earlier.
             i = intersect((abase, abase + alen), (bbase, bbase + blen))
             if i:
                 intbase = i[0]
                 intend = i[1]
                 intlen = intend - intbase
 
                 # found a match of base[i[0], i[1]]; this may be less than
                 # the region that matches in either one
                 assert intlen <= alen
                 assert intlen <= blen
                 assert abase <= intbase
                 assert bbase <= intbase
 
                 asub = amatch + (intbase - abase)
                 bsub = bmatch + (intbase - bbase)
                 aend = asub + intlen
                 bend = bsub + intlen
 
                 assert self.base[intbase:intend] == self.a[asub:aend], (
                     self.base[intbase:intend],
                     self.a[asub:aend],
                 )
 
                 assert self.base[intbase:intend] == self.b[bsub:bend]
 
                 sl.append((intbase, intend, asub, aend, bsub, bend))
 
             # advance whichever one ends first in the base text
             if (abase + alen) < (bbase + blen):
                 ia += 1
             else:
                 ib += 1
 
         intbase = len(self.base)
         abase = len(self.a)
         bbase = len(self.b)
         sl.append((intbase, intbase, abase, abase, bbase, bbase))
 
         return sl
 
 
 def _verifytext(text, path, ui, opts):
     """verifies that text is non-binary (unless opts[text] is passed,
     then we just warn)"""
     if stringutil.binary(text):
         msg = _(b"%s looks like a binary file.") % path
         if not opts.get('quiet'):
             ui.warn(_(b'warning: %s\n') % msg)
         if not opts.get('text'):
             raise error.Abort(msg)
     return text
 
 
 def _picklabels(defaults, overrides):
     if len(overrides) > 3:
         raise error.Abort(_(b"can only specify three labels."))
     result = defaults[:]
     for i, override in enumerate(overrides):
         result[i] = override
     return result
 
 
 def is_not_null(ctx):
     if not util.safehasattr(ctx, "node"):
         return False
-    return ctx.node() != nullid
+    return ctx.rev() != nullrev
 
 
 def _mergediff(m3, name_a, name_b, name_base):
     lines = []
     conflicts = False
     for group in m3.merge_groups():
         if group[0] == b'conflict':
             base_lines, a_lines, b_lines = group[1:]
             base_text = b''.join(base_lines)
             b_blocks = list(
                 mdiff.allblocks(
                     base_text,
                     b''.join(b_lines),
                     lines1=base_lines,
                     lines2=b_lines,
                 )
             )
             a_blocks = list(
                 mdiff.allblocks(
                     base_text,
                     b''.join(a_lines),
                     lines1=base_lines,
                     lines2=b_lines,
                 )
             )
 
             def matching_lines(blocks):
                 return sum(
                     block[1] - block[0]
                     for block, kind in blocks
                     if kind == b'='
                 )
 
             def diff_lines(blocks, lines1, lines2):
                 for block, kind in blocks:
                     if kind == b'=':
                         for line in lines1[block[0] : block[1]]:
                             yield b' ' + line
                     else:
                         for line in lines1[block[0] : block[1]]:
                             yield b'-' + line
                         for line in lines2[block[2] : block[3]]:
                             yield b'+' + line
 
             lines.append(b"<<<<<<<\n")
             if matching_lines(a_blocks) < matching_lines(b_blocks):
                 lines.append(b"======= %s\n" % name_a)
                 lines.extend(a_lines)
                 lines.append(b"------- %s\n" % name_base)
                 lines.append(b"+++++++ %s\n" % name_b)
                 lines.extend(diff_lines(b_blocks, base_lines, b_lines))
             else:
                 lines.append(b"------- %s\n" % name_base)
                 lines.append(b"+++++++ %s\n" % name_a)
                 lines.extend(diff_lines(a_blocks, base_lines, a_lines))
                 lines.append(b"======= %s\n" % name_b)
                 lines.extend(b_lines)
             lines.append(b">>>>>>>\n")
             conflicts = True
         else:
             lines.extend(group[1])
     return lines, conflicts
 
 
 def simplemerge(ui, localctx, basectx, otherctx, **opts):
     """Performs the simplemerge algorithm.
 
     The merged result is written into `localctx`.
     """
 
     def readctx(ctx):
         # Merges were always run in the working copy before, which means
         # they used decoded data, if the user defined any repository
         # filters.
         #
         # Maintain that behavior today for BC, though perhaps in the future
         # it'd be worth considering whether merging encoded data (what the
         # repository usually sees) might be more useful.
         return _verifytext(ctx.decodeddata(), ctx.path(), ui, opts)
 
     mode = opts.get('mode', b'merge')
     name_a, name_b, name_base = None, None, None
     if mode != b'union':
         name_a, name_b, name_base = _picklabels(
             [localctx.path(), otherctx.path(), None], opts.get('label', [])
         )
 
     try:
         localtext = readctx(localctx)
         basetext = readctx(basectx)
         othertext = readctx(otherctx)
     except error.Abort:
         return 1
 
     m3 = Merge3Text(basetext, localtext, othertext)
     extrakwargs = {
         b"localorother": opts.get("localorother", None),
         b'minimize': True,
     }
     if mode == b'union':
         extrakwargs[b'start_marker'] = None
         extrakwargs[b'mid_marker'] = None
         extrakwargs[b'end_marker'] = None
     elif name_base is not None:
         extrakwargs[b'base_marker'] = b'|||||||'
         extrakwargs[b'name_base'] = name_base
         extrakwargs[b'minimize'] = False
 
     if mode == b'mergediff':
         lines, conflicts = _mergediff(m3, name_a, name_b, name_base)
     else:
         lines = list(
             m3.merge_lines(
                 name_a=name_a, name_b=name_b, **pycompat.strkwargs(extrakwargs)
             )
         )
         conflicts = m3.conflicts
 
     # merge flags if necessary
     flags = localctx.flags()
     localflags = set(pycompat.iterbytestr(flags))
     otherflags = set(pycompat.iterbytestr(otherctx.flags()))
     if is_not_null(basectx) and localflags != otherflags:
         baseflags = set(pycompat.iterbytestr(basectx.flags()))
         commonflags = localflags & otherflags
         addedflags = (localflags ^ otherflags) - baseflags
         flags = b''.join(sorted(commonflags | addedflags))
 
     mergedtext = b''.join(lines)
     if opts.get('print'):
         ui.fout.write(mergedtext)
     else:
         localctx.write(mergedtext, flags)
 
     if conflicts and not mode == b'union':
         return 1